© 2006 by Taylor and Francis Group, LLC

Fundamentals of
Natural Gas
Processing
© 2006 by Taylor and Francis Group, LLC
MECHANICAL ENGINEERING
A Series of Textbooks and Reference Books
Founding Editor
L. L. Faulkner
Columbus Division, Battelle Memorial Institute
and Department of Mechanical Engineering
The Ohio State University
Columbus, Ohio
1. Spring Designer’s Handbook, Harold Carlson
2. Computer-Aided Graphics and Design, Daniel L. Ryan
3. Lubrication Fundamentals, J. George Wills
4. Solar Engineering for Domestic Buildings, William A. Himmelman
5. Applied Engineering Mechanics: Statics and Dynamics, G. Boothroyd
and C. Poli
6. Centrifugal Pump Clinic, Igor J. Karassik
7. Computer-Aided Kinetics for Machine Design, Daniel L. Ryan
8. Plastics Products Design Handbook, Part A: Materials
and Components; Part B: Processes and Design for Processes,
edited by Edward Miller
9. Turbomachinery: Basic Theory and Applications, Earl Logan, Jr.
10. Vibrations of Shells and Plates, Werner Soedel
11. Flat and Corrugated Diaphragm Design Handbook, Mario Di Giovanni
12. Practical Stress Analysis in Engineering Design, Alexander Blake
13. An Introduction to the Design and Behavior of Bolted Joints,
John H. Bickford
14. Optimal Engineering Design: Principles and Applications,
James N. Siddall
15. Spring Manufacturing Handbook, Harold Carlson
16. Industrial Noise Control: Fundamentals and Applications,
edited by Lewis H. Bell
17. Gears and Their Vibration: A Basic Approach to Understanding Gear
Noise, J. Derek Smith
18. Chains for Power Transmission and Material Handling:
Design and Applications Handbook, American Chain Association
19. Corrosion and Corrosion Protection Handbook, edited by
Philip A. Schweitzer
20. Gear Drive Systems: Design and Application, Peter Lynwander
21. Controlling In-Plant Airborne Contaminants: Systems Design
and Calculations, John D. Constance
22. CAD/CAM Systems Planning and Implementation, Charles S. Knox
23. Probabilistic Engineering Design: Principles and Applications,
James N. Siddall
© 2006 by Taylor and Francis Group, LLC
24. Traction Drives: Selection and Application, Frederick W. Heilich III
and Eugene E. Shube
25. Finite Element Methods: An Introduction, Ronald L. Huston
and Chris E. Passerello
26. Mechanical Fastening of Plastics: An Engineering Handbook,
Brayton Lincoln, Kenneth J. Gomes, and James F. Braden
27. Lubrication in Practice: Second Edition, edited by W. S. Robertson
28. Principles of Automated Drafting, Daniel L. Ryan
29. Practical Seal Design, edited by Leonard J. Martini
30. Engineering Documentation for CAD/CAM Applications, Charles S. Knox
31. Design Dimensioning with Computer Graphics Applications,
Jerome C. Lange
32. Mechanism Analysis: Simplified Graphical and Analytical Techniques,
Lyndon O. Barton
33. CAD/CAM Systems: Justification, Implementation, Productivity
Measurement, Edward J. Preston, George W. Crawford,
and Mark E. Coticchia
34. Steam Plant Calculations Manual, V. Ganapathy
35. Design Assurance for Engineers and Managers, John A. Burgess
36. Heat Transfer Fluids and Systems for Process and Energy Applications,
Jasbir Singh
37. Potential Flows: Computer Graphic Solutions, Robert H. Kirchhoff
38. Computer-Aided Graphics and Design: Second Edition, Daniel L. Ryan
39. Electronically Controlled Proportional Valves: Selection
and Application, Michael J. Tonyan, edited by Tobi Goldoftas
40. Pressure Gauge Handbook, AMETEK, U.S. Gauge Division,
edited by Philip W. Harland
41. Fabric Filtration for Combustion Sources: Fundamentals and Basic
Technology, R. P. Donovan
42. Design of Mechanical Joints, Alexander Blake
43. CAD/CAM Dictionary, Edward J. Preston, George W. Crawford,
and Mark E. Coticchia
44. Machinery Adhesives for Locking, Retaining, and Sealing,
Girard S. Haviland
45. Couplings and Joints: Design, Selection, and Application, Jon R. Mancuso
46. Shaft Alignment Handbook, John Piotrowski
47. BASIC Programs for Steam Plant Engineers: Boilers, Combustion,
Fluid Flow, and Heat Transfer, V. Ganapathy
48. Solving Mechanical Design Problems with Computer Graphics,
Jerome C. Lange
49. Plastics Gearing: Selection and Application, Clifford E. Adams
50. Clutches and Brakes: Design and Selection, William C. Orthwein
51. Transducers in Mechanical and Electronic Design, Harry L. Trietley
52. Metallurgical Applications of Shock-Wave and High-Strain-Rate
Phenomena, edited by Lawrence E. Murr, Karl P. Staudhammer,
and Marc A. Meyers
53. Magnesium Products Design, Robert S. Busk
54. How to Integrate CAD/CAM Systems: Management and Technology,
William D. Engelke
© 2006 by Taylor and Francis Group, LLC
55. Cam Design and Manufacture: Second Edition; with cam design software
for the IBM PC and compatibles, disk included, Preben W. Jensen
56. Solid-State AC Motor Controls: Selection and Application,
Sylvester Campbell
57. Fundamentals of Robotics, David D. Ardayfio
58. Belt Selection and Application for Engineers, edited by
Wallace D. Erickson
59. Developing Three-Dimensional CAD Software with the IBM PC,
C. Stan Wei
60. Organizing Data for CIM Applications, Charles S. Knox, with contributions
by Thomas C. Boos, Ross S. Culverhouse, and Paul F. Muchnicki
61. Computer-Aided Simulation in Railway Dynamics, by Rao V. Dukkipati
and Joseph R. Amyot
62. Fiber-Reinforced Composites: Materials, Manufacturing, and Design,
P. K. Mallick
63. Photoelectric Sensors and Controls: Selection and Application,
Scott M. Juds
64. Finite Element Analysis with Personal Computers, Edward R. Champion,
Jr. and J. Michael Ensminger
65. Ultrasonics: Fundamentals, Technology, Applications: Second Edition,
Revised and Expanded, Dale Ensminger
66. Applied Finite Element Modeling: Practical Problem Solving for Engineers,
Jeffrey M. Steele
67. Measurement and Instrumentation in Engineering: Principles and Basic
Laboratory Experiments, Francis S. Tse and Ivan E. Morse
68. Centrifugal Pump Clinic: Second Edition, Revised and Expanded,
Igor J. Karassik
69. Practical Stress Analysis in Engineering Design: Second Edition,
Revised and Expanded, Alexander Blake
70. An Introduction to the Design and Behavior of Bolted Joints:
Second Edition, Revised and Expanded, John H. Bickford
71. High Vacuum Technology: A Practical Guide, Marsbed H. Hablanian
72. Pressure Sensors: Selection and Application, Duane Tandeske
73. Zinc Handbook: Properties, Processing, and Use in Design, Frank Porter
74. Thermal Fatigue of Metals, Andrzej Weronski and Tadeusz Hejwowski
75. Classical and Modern Mechanisms for Engineers and Inventors,
Preben W. Jensen
76. Handbook of Electronic Package Design, edited by Michael Pecht
77. Shock-Wave and High-Strain-Rate Phenomena in Materials, edited by
Marc A. Meyers, Lawrence E. Murr, and Karl P. Staudhammer
78. Industrial Refrigeration: Principles, Design and Applications, P. C. Koelet
79. Applied Combustion, Eugene L. Keating
80. Engine Oils and Automotive Lubrication, edited by Wilfried J. Bartz
81. Mechanism Analysis: Simplified and Graphical Techniques, Second
Edition, Revised and Expanded, Lyndon O. Barton
82. Fundamental Fluid Mechanics for the Practicing Engineer,
James W. Murdock
83. Fiber-Reinforced Composites: Materials, Manufacturing, and Design,
Second Edition, Revised and Expanded, P. K. Mallick
© 2006 by Taylor and Francis Group, LLC
84. Numerical Methods for Engineering Applications,
Edward R. Champion, Jr.
85. Turbomachinery: Basic Theory and Applications, Second Edition,
Revised and Expanded, Earl Logan, Jr.
86. Vibrations of Shells and Plates: Second Edition, Revised and Expanded,
Werner Soedel
87. Steam Plant Calculations Manual: Second Edition, Revised
and Expanded, V. Ganapathy
88. Industrial Noise Control: Fundamentals and Applications, Second Edition,
Revised and Expanded, Lewis H. Bell and Douglas H. Bell
89. Finite Elements: Their Design and Performance, Richard H. MacNeal
90. Mechanical Properties of Polymers and Composites:
Second Edition, Revised and Expanded, Lawrence E. Nielsen
and Robert F. Landel
91. Mechanical Wear Prediction and Prevention, Raymond G. Bayer
92. Mechanical Power Transmission Components, edited by
David W. South and Jon R. Mancuso
93. Handbook of Turbomachinery, edited by Earl Logan, Jr.
94. Engineering Documentation Control Practices and Procedures,
Ray E. Monahan
95. Refractory Linings Thermomechanical Design and Applications,
Charles A. Schacht
96. Geometric Dimensioning and Tolerancing: Applications and Techniques
for Use in Design, Manufacturing,
and Inspection, James D. Meadows
97. An Introduction to the Design and Behavior of Bolted Joints: Third Edition,
Revised and Expanded, John H. Bickford
98. Shaft Alignment Handbook: Second Edition, Revised and Expanded,
John Piotrowski
99. Computer-Aided Design of Polymer-Matrix Composite Structures,
edited by Suong Van Hoa
100. Friction Science and Technology, Peter J. Blau
101. Introduction to Plastics and Composites: Mechanical Properties
and Engineering Applications, Edward Miller
102. Practical Fracture Mechanics in Design, Alexander Blake
103. Pump Characteristics and Applications, Michael W. Volk
104. Optical Principles and Technology for Engineers, James E. Stewart
105. Optimizing the Shape of Mechanical Elements and Structures,
A. A. Seireg and Jorge Rodriguez
106. Kinematics and Dynamics of Machinery, Vladimír Stejskal
and Michael Valásek
107. Shaft Seals for Dynamic Applications, Les Horve
108. Reliability-Based Mechanical Design, edited by Thomas A. Cruse
109. Mechanical Fastening, Joining, and Assembly, James A. Speck
110. Turbomachinery Fluid Dynamics and Heat Transfer, edited by Chunill Hah
111. High-Vacuum Technology: A Practical Guide, Second Edition,
Revised and Expanded, Marsbed H. Hablanian
112. Geometric Dimensioning and Tolerancing: Workbook and Answerbook,
James D. Meadows
© 2006 by Taylor and Francis Group, LLC
113. Handbook of Materials Selection for Engineering Applications,
edited by G. T. Murray
114. Handbook of Thermoplastic Piping System Design, Thomas Sixsmith
and Reinhard Hanselka
115. Practical Guide to Finite Elements: A Solid Mechanics Approach,
Steven M. Lepi
116. Applied Computational Fluid Dynamics, edited by Vijay K. Garg
117. Fluid Sealing Technology, Heinz K. Muller and Bernard S. Nau
118. Friction and Lubrication in Mechanical Design, A. A. Seireg
119. Influence Functions and Matrices, Yuri A. Melnikov
120. Mechanical Analysis of Electronic Packaging Systems,
Stephen A. McKeown
121. Couplings and Joints: Design, Selection, and Application, Second Edition,
Revised and Expanded, Jon R. Mancuso
122. Thermodynamics: Processes and Applications, Earl Logan, Jr.
123. Gear Noise and Vibration, J. Derek Smith
124. Practical Fluid Mechanics for Engineering Applications, John J. Bloomer
125. Handbook of Hydraulic Fluid Technology, edited by George E. Totten
126. Heat Exchanger Design Handbook, T. Kuppan
127. Designing for Product Sound Quality, Richard H. Lyon
128. Probability Applications in Mechanical Design, Franklin E. Fisher
and Joy R. Fisher
129. Nickel Alloys, edited by Ulrich Heubner
130. Rotating Machinery Vibration: Problem Analysis and Troubleshooting,
Maurice L. Adams, Jr.
131. Formulas for Dynamic Analysis, Ronald L. Huston and C. Q. Liu
132. Handbook of Machinery Dynamics, Lynn L. Faulkner and Earl Logan, Jr.
133. Rapid Prototyping Technology: Selection and Application,
Kenneth G. Cooper
134. Reciprocating Machinery Dynamics: Design and Analysis,
Abdulla S. Rangwala
135. Maintenance Excellence: Optimizing Equipment Life-Cycle Decisions,
edited by John D. Campbell and Andrew K. S. Jardine
136. Practical Guide to Industrial Boiler Systems, Ralph L. Vandagriff
137. Lubrication Fundamentals: Second Edition, Revised and Expanded,
D. M. Pirro and A. A. Wessol
138. Mechanical Life Cycle Handbook: Good Environmental Design
and Manufacturing, edited by Mahendra S. Hundal
139. Micromachining of Engineering Materials, edited by Joseph McGeough
140. Control Strategies for Dynamic Systems: Design and Implementation,
John H. Lumkes, Jr.
141. Practical Guide to Pressure Vessel Manufacturing, Sunil Pullarcot
142. Nondestructive Evaluation: Theory, Techniques, and Applications,
edited by Peter J. Shull
143. Diesel Engine Engineering: Thermodynamics, Dynamics, Design,
and Control, Andrei Makartchouk
144. Handbook of Machine Tool Analysis, Ioan D. Marinescu, Constantin Ispas,
and Dan Boboc
© 2006 by Taylor and Francis Group, LLC
145. Implementing Concurrent Engineering in Small Companies,
Susan Carlson Skalak
146. Practical Guide to the Packaging of Electronics: Thermal and Mechanical
Design and Analysis, Ali Jamnia
147. Bearing Design in Machinery: Engineering Tribology and Lubrication,
Avraham Harnoy
148. Mechanical Reliability Improvement: Probability and Statistics for
Experimental Testing, R. E. Little
149. Industrial Boilers and Heat Recovery Steam Generators: Design,
Applications, and Calculations, V. Ganapathy
150. The CAD Guidebook: A Basic Manual for Understanding
and Improving Computer-Aided Design, Stephen J. Schoonmaker
151. Industrial Noise Control and Acoustics, Randall F. Barron
152. Mechanical Properties of Engineered Materials, Wolé Soboyejo
153. Reliability Verification, Testing, and Analysis in Engineering Design,
Gary S. Wasserman
154. Fundamental Mechanics of Fluids: Third Edition, I. G. Currie
155. Intermediate Heat Transfer, Kau-Fui Vincent Wong
156. HVAC Water Chillers and Cooling Towers: Fundamentals, Application,
and Operation, Herbert W. Stanford III
157. Gear Noise and Vibration: Second Edition, Revised and Expanded,
J. Derek Smith
158. Handbook of Turbomachinery: Second Edition, Revised and Expanded,
edited by Earl Logan, Jr. and Ramendra Roy
159. Piping and Pipeline Engineering: Design, Construction, Maintenance,
Integrity, and Repair, George A. Antaki
160. Turbomachinery: Design and Theory, Rama S. R. Gorla
and Aijaz Ahmed Khan
161. Target Costing: Market-Driven Product Design, M. Bradford Clifton,
Henry M. B. Bird, Robert E. Albano, and Wesley P. Townsend
162. Fluidized Bed Combustion, Simeon N. Oka
163. Theory of Dimensioning: An Introduction to Parameterizing Geometric
Models, Vijay Srinivasan
164. Handbook of Mechanical Alloy Design, edited by George E. Totten,
Lin Xie, and Kiyoshi Funatani
165. Structural Analysis of Polymeric Composite Materials, Mark E. Tuttle
166. Modeling and Simulation for Material Selection and Mechanical Design,
edited by George E. Totten, Lin Xie, and Kiyoshi Funatani
167. Handbook of Pneumatic Conveying Engineering, David Mills,
Mark G. Jones, and Vijay K. Agarwal
168. Clutches and Brakes: Design and Selection, Second Edition,
William C. Orthwein
169. Fundamentals of Fluid Film Lubrication: Second Edition,
Bernard J. Hamrock, Steven R. Schmid, and Bo O. Jacobson
170. Handbook of Lead-Free Solder Technology for Microelectronic
Assemblies, edited by Karl J. Puttlitz and Kathleen A. Stalter
171. Vehicle Stability, Dean Karnopp
172. Mechanical Wear Fundamentals and Testing: Second Edition,
Revised and Expanded, Raymond G. Bayer
173. Liquid Pipeline Hydraulics, E. Shashi Menon
© 2006 by Taylor and Francis Group, LLC
174. Solid Fuels Combustion and Gasification, Marcio L. de Souza-Santos
175. Mechanical Tolerance Stackup and Analysis, Bryan R. Fischer
176. Engineering Design for Wear, Raymond G. Bayer
177. Vibrations of Shells and Plates: Third Edition, Revised and Expanded,
Werner Soedel
178. Refractories Handbook, edited by Charles A. Schacht
179. Practical Engineering Failure Analysis, Hani M. Tawancy, Anwar Ul-Hamid,
and Nureddin M. Abbas
180. Mechanical Alloying and Milling, C. Suryanarayana
181. Mechanical Vibration: Analysis, Uncertainties, and Control,
Second Edition, Revised and Expanded, Haym Benaroya
182. Design of Automatic Machinery, Stephen J. Derby
183. Practical Fracture Mechanics in Design: Second Edition,
Revised and Expanded, Arun Shukla
184. Practical Guide to Designed Experiments, Paul D. Funkenbusch
185. Gigacycle Fatigue in Mechanical Practive, Claude Bathias
and Paul C. Paris
186. Selection of Engineering Materials and Adhesives, Lawrence W. Fisher
187. Boundary Methods: Elements, Contours, and Nodes, Subrata Mukherjee
and Yu Xie Mukherjee
188. Rotordynamics, Agnieszka (Agnes) Muszn´yska
189. Pump Characteristics and Applications: Second Edition, Michael W. Volk
190. Reliability Engineering: Probability Models and Maintenance Methods,
Joel A. Nachlas
191. Industrial Heating: Principles, Techniques, Materials, Applications,
and Design, Yeshvant V. Deshmukh
192. Micro Electro Mechanical System Design, James J. Allen
193. Probability Models in Engineering and Science, Haym Benaroya
and Seon Han
194. Damage Mechanics, George Z. Voyiadjis and Peter I. Kattan
195. Standard Handbook of Chains: Chains for Power Transmission
and Material Handling, Second Edition, American Chain Association
and John L. Wright, Technical Consultant
196. Standards for Engineering Design and Manufacturing,
Wasim Ahmed Khan and Abdul Raouf S.I.
197. Maintenance, Replacement, and Reliability: Theory and Applications,
Andrew K. S. Jardine and Albert H. C. Tsang
198. Finite Element Method: Applications in Solids, Structures, and Heat
Transfer, Michael R. Gosz
199. Microengineering, MEMS, and Interfacing: A Practical Guide,
Danny Banks
200. Fundamentals of Natural Gas Processing, Arthur J. Kidnay
and William Parrish
© 2006 by Taylor and Francis Group, LLC
Fundamentals of
Natural Gas
Processing
Arthur J. Kidnay
William R. Parrish
CRC is an imprint of the Taylor & Francis Group,
an informa business
Boca Raton London New York
© 2006 by Taylor and Francis Group, LLC
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
© 2006 by Taylor and Francis Group, LLC
CRC Press is an imprint of Taylor & Francis Group, an Informa business
No claim to original U.S. Government works
Printed in the United States of America on acid-free paper
10 9 8 7 6 5 4 3 2 1
International Standard Book Number-10: 0-8493-3406-3 (Hardcover)
International Standard Book Number-13: 978-0-8493-3406-1 (Hardcover)
Library of Congress Card Number 2005036359
This book contains information obtained from authentic and highly regarded sources. Reprinted mate-
rial is quoted with permission, and sources are indicated. A wide variety of references are listed. Reason-
able efforts have been made to publish reliable data and information, but the author and the publisher
cannot assume responsibility for the validity of all materials or for the consequences of their use.
No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any elec-
tronic, mechanical, or other means, now known or hereafter invented, including photocopying, micro-
filming, and recording, or in any information storage or retrieval system, without written permission
from the publishers.
Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides
licenses and registration for a variety of users. For organizations that have been granted a photocopy
license by the CCC, a separate system of payment has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are
used only for identification and explanation without intent to infringe.
Library of Congress Cataloging-in-Publication Data
Kidnay, A. J.
Fundamentals of natural gas processing / Arthur J. Kidnay and William Parrish.
p. cm. -- (Mechanical engineering)
Includes bibliographical references and index.
ISBN-13: 978-0-8493-3406-1 (acid-free paper)
ISBN-10: 0-8493-3406-3 (acid-free paper)
1. Gas industry. I. Parrish, William, 1914- II. Title. III. Mechanical engineering
series (Boca Raton, Fla.)
TP751.K54 2006
665.7’3--dc22 2005036359
Visit the Taylor & Francis Web site at
and the CRC Press Web site at
For permission to photocopy or use material electronically from this work, please access www.copy-
right.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC) 222
http://www.taylorandfrancis.com
http://www.crcpress.com
© 2006 by Taylor and Francis Group, LLC
Dedication
To our wives, Joan and Joan, for their enduring support
and patience throughout the preparation of this book.
© 2006 by Taylor and Francis Group, LLC
Preface
The natural gas industry began in the early 1900s in the United States and is still
evolving. This high-quality fuel and chemical feedstock plays an important role
in the industrial world and is becoming an important export for other countries.
Several high-quality books* provide guidance to those experienced in natural gas
processing. This book introduces the natural gas industry to a reader entering the
field. It also helps those providing a service to the industry in a narrow application
to better understand how their products and services fit into the overall process.
To help the reader understand the need of each processing step, the book
follows the gas stream from the wellhead to the market place. The book focuses
primarily on the gas plant processes. Wherever possible, the advantages, limita-
tions, and ranges of applicability of the processes are discussed so that their
selection and integration into the overall gas plant can be fully understood and
appreciated.
The book compiles information from other books, open literature, and meet-
ing proceedings** to hopefully give an accurate picture of where the gas pro-
cessing technology stands today, as well as indicate some relatively new tech-
nologies that could become important in the future. An invaluable contribution
to the book is the insight provided to the authors by experts in certain applications.
* For example, GPSA Engineering Data Book (Gas Processors Supply Association, Tulsa, OK, 12th
Edition, 2004), and the fifth edition of Kohl and Nielsen, Gas Purification (Gulf Publishing, Houston,
TX, 1997).
** The two most important meetings involving natural gas processing in the United States are the
annual meeting of the Gas Processors Association and the Laurance Reid Gas Conditioning
Conference.
© 2006 by Taylor and Francis Group, LLC
Acknowledgments
The authors communicated with numerous people in preparing this book. It could
not have been written without the aid of the Gas Processors Association (GPA).
Ron Brunner graciously supplied requested material from the vast literature on
gas processing available through the GPA. Dan McCartney provided valuable
insight and comments while he generously took time to review the manuscript.
In most cases, the private communications referenced in this book involved
numerous letters and conversations. Phil Richman and John Peranteaux willingly
provided both technical input and editorial comment. Others who provided valu-
able input include Joe Kuchinski, Charles Wallace, Ed Wichert, Dendy Sloan,
Veet Kruka, and Dale Embry. A number of companies graciously provided us
with drawings and photographs. One company generously supplied a modified
drawing that replaced their product names with generic names so that the figure
could be used. Finally, we appreciate the patience and assistance of the editorial
staff at Taylor and Francis.
Carter Tannehill kindly provided us with the cost data provided in Chapter 14.
© 2006 by Taylor and Francis Group, LLC
AUTHORS
Arthur Kidnay, Ph.D., P. E., is professor emeritus, Chemical Engineering
Department, Colorado School of Mines (CSM). He was a research engineer with
the National Institute of Standards and Technology (NIST) for 9 years before
joining the faculty of CSM. He has taught and conducted extensive research in
the fields of vapor−liquid equilibria, physical adsorption, and heat transfer. Dr.
Kidnay is the author of 69 technical papers and has advised 42 M.S. and Ph.D.
students. He remains very active in professional activities at CSM and presently
teaches a senior course in natural gas processing. For 26 years, Dr. Kidnay and
four colleagues have taught a continuing education course in gas processing to
engineers and scientists from the natural gas industry.
In recognition of his services to the engineering profession, he was elected
a Fellow of the American Institute of Chemical Engineers, in 1987 and was
appointed by the governor of Colorado to two terms (1984−1992) on the Board
of Registration for Professional Engineers. He served on the Cryogenic Confer-
ence Executive Board from 1969 through 1972 and received the Russell B. Scott
Memorial Award for the outstanding technical paper presented at the 1966 Cryo-
genic Engineering Conference. Professor Kidnay was NATO Senior Science
Fellow at Oxford University in the summer of 1972.
William R Parrish, Ph.D., P.E., is a retired senior research associate. He spent
25 years in research and development at ConocoPhillips (formerly Phillips Petro-
leum Company) where he obtained physical properties data needed for new
processes and for resolving operation problems. He provided company-wide
technical expertise on matters involving physical properties and gas hydrates. He
also participated on six gas plant optimization teams. His work has appeared in
49 technical publications and he holds two patents. He presently teaches a con-
tinuing education course in gas processing for engineers and scientists from
industry.
Dr. Parrish represented his company on various committees including the Gas
Processors Association’s Enthalpy Committee of Section F. He also participated
on Department of Energy peer review committees. He is a Fellow of the American
Institute of Chemical Engineers and is actively involved in professional engineer
examination development.
© 2006 by Taylor and Francis Group, LLC
Table of Contents
Chapter 1 Overview of Natural Gas Industry..................................................1
1.1 Introduction..................................................................................................1
1.1.1 The World Picture for Natural Gas.................................................2
1.1.2 Natural Gas in United States...........................................................5
1.1.3 Nonconventional Gas Reserves in United States............................7
1.2 Sources of Natural Gas ...............................................................................7
1.3 Natural Gas Compositions ..........................................................................9
1.3.1 Traditional Natural Gas...................................................................9
1.3.2 Important Impurities......................................................................10
1.3.3 Coal Bed Methane.........................................................................10
1.3.4 Subquality Gas ..............................................................................11
1.4 Classification..............................................................................................11
1.4.1 Liquids Content .............................................................................11
1.4.2 Sulfur Content ...............................................................................13
1.5 Processing and Principal Products ............................................................13
1.5.1 Methane .........................................................................................14
1.5.2 Ethane ............................................................................................14
1.5.3 Propane ..........................................................................................14
1.5.4 Ethane–Propane Mix .....................................................................14
1.5.5 Isobutane........................................................................................15
1.5.6 n-Butane.........................................................................................15
1.5.7 Natural Gas Liquids ......................................................................15
1.5.8 Natural Gasoline............................................................................15
1.5.9 Sulfur .............................................................................................15
1.6 Product Specifications ...............................................................................16
1.6.1 Natural Gas....................................................................................16
1.6.2 Liquid Products .............................................................................17
1.7 Combustion Characteristics.......................................................................18
1.7.1 Heating Value ................................................................................18
1.7.2 Wobbe Number..............................................................................20
References ...........................................................................................................21
Chapter 2 Overview of Gas Plant Processing................................................25
2.1 Roles of Gas Plants ...................................................................................25
2.2 Plant Processes ..........................................................................................26
2.2.1 Field Operations and Inlet Receiving ...........................................26
2.2.2 Inlet Compression..........................................................................26
2.2.3 Gas Treating ..................................................................................26
© 2006 by Taylor and Francis Group, LLC
2.2.4 Dehydration ...................................................................................27
2.2.5 Hydrocarbon Recovery..................................................................28
2.2.6 Nitrogen Rejection.........................................................................28
2.2.7 Helium Recovery...........................................................................28
2.2.8 Outlet Compression.......................................................................28
2.2.9 Liquids Processing.........................................................................28
2.2.10 Sulfur Recovery.............................................................................28
2.2.11 Storage and Transportation ...........................................................29
2.2.12 Liquefaction...................................................................................29
2.3 Important Support Components ................................................................29
2.3.1 Utilities ..........................................................................................29
2.3.2 Process Control..............................................................................30
2.3.3 Safety Systems...............................................................................30
2.4 Contractual Agreements and Economics ..................................................30
2.4.1 Fee-Based Contracts......................................................................31
2.4.2 Percentage of Proceeds Contracts.................................................31
2.4.3 Wellhead Purchase Contracts ........................................................31
2.4.4 Fixed Efficiency Contracts ............................................................31
2.4.5 Keep Whole Contracts...................................................................31
References ...........................................................................................................32
Chapter 3 Field Operations and Inlet Receiving ...........................................33
3.1 Introduction................................................................................................33
3.2 Field Operations ........................................................................................34
3.2.1 Wellhead Operations .....................................................................34
3.2.2 Piping.............................................................................................35
3.2.3 Compression Stations ....................................................................36
3.2.4 Pigging...........................................................................................38
3.3 Gas Hydrates .............................................................................................40
3.3.1 Properties .......................................................................................40
3.3.2 Hydrate Formation Prediction.......................................................42
3.3.3 Hydrate Inhibition .........................................................................44
3.4 Inlet Receiving...........................................................................................49
3.4.1 Separator Principles.......................................................................49
3.4.2 Slug Catcher Configurations .........................................................56
3.5 Safety and Environmental Considerations................................................60
References ...........................................................................................................61
Chapter 4 Compression ..................................................................................63
4.1 Introduction................................................................................................63
4.2 Fundamentals.............................................................................................65
4.2.1 Thermodynamics of Compression ................................................65
4.2.2 Multistaging...................................................................................68
4.2.3 Compressor Efficiencies................................................................69
© 2006 by Taylor and Francis Group, LLC
4.3 Compressor Types .....................................................................................71
4.3.1 Positive Displacement Compressors .............................................72
4.3.2 Dynamic Compressors...................................................................76
4.4 Capacity and Power Calculations .............................................................81
4.4.1 Capacity.........................................................................................82
4.4.2 Power Requirements......................................................................85
4.5 Comparison of Reciprocating and Centrifugal
Compressors ..............................................................................................87
4.6 Safety and Environmental Considerations................................................88
References ...........................................................................................................89
Chapter 5 Gas Treating ..................................................................................91
5.1 Introduction................................................................................................91
5.1.1 The Problem..................................................................................92
5.1.2 Acid Gas Concentrations in Natural Gas......................................92
5.1.3 Purification Levels.........................................................................93
5.1.4 Acid Gas Disposal.........................................................................93
5.1.5 Purification Processes....................................................................94
5.2 Solvent Absorption Processes ...................................................................99
5.2.1 Amines.........................................................................................100
5.2.2 Alkali Salts ..................................................................................109
5.3 Physical Absorption.................................................................................110
5.3.1 Solvent Properties........................................................................110
5.3.2 Representative Process Conditions .............................................112
5.3.3 Hybrid Processes .........................................................................114
5.4 Adsorption ...............................................................................................115
5.5 Cryogenic Fractionation..........................................................................117
5.6 Membranes ..............................................................................................119
5.6.1 Membrane Fundamentals ............................................................119
5.6.2 Carbon Dioxide Removal from Natural Gas ..............................121
5.6.3 Operating Considerations ............................................................123
5.6.4 Advantages and Disadvantages of Membrane Systems .............126
5.7 Nonregenerable Hydrogen Sulfide Scavengers.......................................127
5.8 Biological Processes................................................................................128
5.9 Safety and Environmental Considerations..............................................129
5.9.1 Amines.........................................................................................129
5.9.2 Physical Absorption.....................................................................129
5.9.3 Adsorption ...................................................................................129
5.9.4 Membranes ..................................................................................129
References .........................................................................................................130
Chapter 6 Gas Dehydration..........................................................................133
6.1 Introduction..............................................................................................133
6.2 Water Content of Hydrocarbons .............................................................134
© 2006 by Taylor and Francis Group, LLC
6.3 Gas Dehydration Processes .....................................................................138
6.3.1 Absorption Processes...................................................................139
6.3.2 Adsorption Processes...................................................................146
6.3.3 Desiccant Processes.....................................................................160
6.3.4 Membrane Processes ...................................................................160
6.3.5 Other Processes ...........................................................................161
6.3.6 Comparison of Dehydration Processes .......................................161
6.4 Safety and Environmental Considerations..............................................162
References .........................................................................................................162
Chapter 7 Hydrocarbon Recovery................................................................165
7.1 Introduction..............................................................................................165
7.1.1 Retrograde Condensation ............................................................166
7.2 Process Components................................................................................167
7.2.1 External Refrigeration .................................................................168
7.2.2 Turboexpansion............................................................................174
7.2.3 Heat Exchange.............................................................................179
7.2.4 Fractionation................................................................................181
7.3 Recovery Processes .................................................................................183
7.3.1 Dew Point Control and Fuel Conditioning.................................184
7.3.2 Lower Ethane Recovery ..............................................................188
7.3.3 High Ethane Recovery ................................................................193
7.4 Safety and Environmental Considerations..............................................196
References .........................................................................................................197
Chapter 8 Nitrogen Rejection.......................................................................199
8.1 Introduction..............................................................................................199
8.2 Nitrogen Rejection for Gas Upgrading...................................................200
8.2.1 Cryogenic Distillation .................................................................201
8.2.2 Pressure Swing Adsorption .........................................................202
8.2.3 Membranes ..................................................................................204
8.3 Nitrogen Rejection for Enhanced Oil Recovery.....................................204
8.4 Safety and Environmental Considerations..............................................206
References .........................................................................................................206
Chapter 9 Trace-Component Recovery or Removal....................................209
9.1 Introduction..............................................................................................209
9.1.1 Hydrogen .....................................................................................210
9.1.2 Oxygen.........................................................................................210
9.1.3 Radon (NORM) ...........................................................................211
9.1.4 Arsenic.........................................................................................211
© 2006 by Taylor and Francis Group, LLC
9.2 Helium .....................................................................................................211
9.2.1 Introduction..................................................................................211
9.2.2 Recovery Methods.......................................................................212
9.3 Mercury....................................................................................................215
9.3.1 Environmental Considerations ....................................................216
9.3.2 Amalgam Formation....................................................................217
9.3.3 Removal Processes ......................................................................217
9.4 (BTEX) Benzene, Toluene, Ethylbenzene, and Xylene .........................218
References .........................................................................................................220
Chapter 10 Liquids Processing.....................................................................223
10.1 Introduction ............................................................................................223
10.2 Condensate Processing...........................................................................224
10.2.1 Sweetening ................................................................................225
10.2.2 Dehydration...............................................................................225
10.3 NGL Processing .....................................................................................225
10.3.1 Sweetening ................................................................................226
10.3.2 Dehydration...............................................................................230
10.3.3 Fractionation..............................................................................233
10.4 Safety and Environmental Considerations.............................................234
References .........................................................................................................235
Chapter 11 Sulfur Recovery .........................................................................237
11.1 Introduction ............................................................................................237
11.2 Properties of Sulfur ................................................................................238
11.3 Sulfur Recovery Processes.....................................................................239
11.3.1 Claus Process ............................................................................239
11.3.2 Claus Tail Gas Cleanup ............................................................242
11.4 Sulfur Storage.........................................................................................246
11.5 Safety and Environmental Considerations.............................................246
References .........................................................................................................248
Chapter 12 Transportation and Storage .......................................................251
12.1 Introduction ............................................................................................251
12.2 Gas..........................................................................................................252
12.2.1 Transportation ...........................................................................252
12.2.2 Market Centers..........................................................................254
12.2.3 Storage.......................................................................................254
12.3 Liquids....................................................................................................259
12.3.1 Transportation ...........................................................................259
12.3.2 Storage.......................................................................................262
References .........................................................................................................263
© 2006 by Taylor and Francis Group, LLC

Chapter 13

Liquefied Natural Gas ...............................................................265
13.1 Introduction ............................................................................................265
13.1.1 Peak Shaving Plants and Satellite Facilities.............................266
13.1.2 Baseload Plants and Stranded Reserves ...................................267
13.2 Gas Treating before Liquefaction ..........................................................270
13.3 Liquefaction Cycles................................................................................272
13.3.1 Joule-Thomson Cycles..............................................................272
13.3.2 Expander Cycles .......................................................................280
13.3.3 Cascade Cycles .........................................................................285
13.4 Storage of LNG......................................................................................292
13.4.1 Cryogenic Aboveground Storage..............................................293
13.4.2 Cryogenic In Ground Storage .................................................296
13.4.3 Rollover .....................................................................................298
13.5 Transportation.........................................................................................300
13.5.1 Truck Transport .........................................................................301
13.5.2 Pipelines ....................................................................................301
13.5.3 Marine Transport.......................................................................301
13.6 Regasification and Cold Utilization of LNG.........................................305
13.6.1 Regasification ............................................................................305
13.6.2 Cold Utilization.........................................................................305
13.7 Economics ..............................................................................................306
13.7.1 Liquefaction Costs ....................................................................306
13.7.2 Shipping Costs ..........................................................................307
13.7.3 Regasification Terminal Costs ..................................................308
13.8 Plant Efficiency ......................................................................................308
13.9 Safety and Environmental Considerations.............................................309
References .........................................................................................................310

Chapter 14

Capital Costs of Gas Processing Facilities ...............................315
14.1 Introduction ............................................................................................315
14.2 Basic Premises for Cost Data ................................................................315
14.3 Amine Treating.......................................................................................315
14.4 Glycol Dehydration................................................................................317
14.5 NGL Recovery with Straight Refrigeration
(Low Ethane Recovery)...........................................................................317
14.6 NGL Recovery with Cryogenic Processing
(High Ethane Recovery)..........................................................................318
14.7 Sulfur Recovery and Tail Gas Cleanup .................................................318
14.7.1 High Sulfur Recovery Rates .....................................................318
14.7.2 Low Sulfur Recovery Rates......................................................319
14.8 NGL Extraction Plant Costs for Larger Facilities.................................321
14.9 Corrections to Cost Data........................................................................323
References .........................................................................................................323

DK063X_C000.fm Page xxiv Thursday, May 18, 2006 3:05 PM
© 2006 by Taylor and Francis Group, LLC
Chapter 15 Natural Gas Processing Plants ..................................................325
15.1 Introduction ............................................................................................325
15.2 Plant with Sweet Gas Feed and 98% Ethane Recovery........................325
15.2.1 Overview of Plant Feed and Product Slate ..............................325
15.2.2 Compression..............................................................................326
15.2.3 Heat Exchange ..........................................................................326
15.2.4 Dehydration...............................................................................326
15.2.5 Propane Refrigeration ...............................................................327
14.2.6 Hydrocarbon Recovery .............................................................328
15.2.7 Amine Treating .........................................................................328
15.2.8 Deethanizer................................................................................328
15.2.9 Residue Compression................................................................328
15.3 Plant with Sour Gas Feed, NGL,
and Sulfur Recovery...............................................................................329
15.3.1 Overview of Plant Feed and Product Slate ..............................329
15.3.2 Inlet Receiving ..........................................................................329
15.3.3 Inlet Compression .....................................................................330
15.3.4 Gas Treating..............................................................................330
15.3.5 Sulfur Recovery ........................................................................330
15.3.6 Dehydration...............................................................................331
15.3.7 Hydrocarbon Recovery .............................................................331
15.3.8 Liquids Processing....................................................................331
15.4 Plant with Sour Gas Feed, NGL Recovery,
and Nitrogen Rejection ..........................................................................332
14.4.1 Overview of Plant Feed and Product Slate ..............................332
14.4.2 Inlet Receiving ..........................................................................332
14.2.3 Gas Treating..............................................................................333
14.2.4 Sulfur Recovery ........................................................................333
14.2.5 Dehydration...............................................................................333
14.2.6 NRU and Cold Box...................................................................334
14.2.7 Liquids Processing....................................................................334
References .........................................................................................................334
Chapter 16 Notation .....................................................................................335
Appendix A Glossary of Gas Process Terminology.....................................339
Appendix B Physical Constants and Physical Properties .............................351
B.1 Unit Conversion Factors..........................................................................354
B.2 Gas Constants and Standard Gas Conditions .........................................355
B.3 Thermodynamic and Physical Property Data ........................................355
B.4 Hydrocarbon Compressibility Factors ....................................................416
References ....................................................................................................... 418
© 2006 by Taylor and Francis Group, LLC
1
1
Overview of the
Natural Gas Industry
1.1 INTRODUCTION
The Chinese are reputed to have been the first to use natural gas commercially,
some 2,400 years ago. The gas was obtained from shallow wells, transported in
bamboo pipes, and used to produce salt from brine in gas-fired evaporators.
Manufactured, or town, gas (gas manufactured from coal) was used in both Britain
and the United States in the late 17
th
and early 18
th
centuries for streetlights and
house lighting The next recorded commercial use of natural gas occurred in 1821.
William Hart drilled a shallow 30-foot (9-meter) well in Fredonia, New York, and,
by use of wooden pipes, transported the gas to local houses and stores (Natural
Gas Suppliers Association, 2004).
During the following years, a number of small, local programs involved natural
gas, but large-scale activity began in the early years of the 20
th
century. The major
boom in gas usage occurred after World War II, when engineering advances allowed
the construction of safe, reliable, long-distance pipelines for gas transportation. At
the end of 2004, the United States had more than 297,000 miles (479,000 kilome-
ters) of gas pipelines, both interstate and intrastate. In 2004 the U.S. was the world’s
second largest producer of natural gas (19.2 trillion cubic feet [Tcf]*, 543 BSm
3
)
and the leading world consumer (22.9 Tcf, 647 BSm
3
). (Energy Information Admin-
istration, 2005h and BP Statistical Review of World Energy, 2005)
Although the primary use of natural gas is as a fuel, it is also a source of
hydrocarbons for petrochemical feedstocks and a major source of elemental sulfur,
an important industrial chemical. Its popularity as an energy source is expected to
grow substantially in the future because natural gas presents many environmental
greenhouse gas linked to global warming, is produced from oil and coal at a rate
approximately 1.4 to 1.75 times higher than production from natural gas.
Both atmospheric nitrogen and nitrogen in fuel are sources of nitrogen oxides
(NO
X
), which are greenhouse gases and a source of acid rain. Because both oil
and coal contain nitrogen compounds not present in natural gas, the nitrogen
oxides formed from burning natural gas are approximately 20% of those produced
* Gas volumes are normally reported in terms of standard cubic feet (scf) at standard conditions of
60°F and 14.7 psia. In metric units, the volumes are given in either normal cubic meters, Nm
3
, where
standard conditions are 0°C, 1 bar, or standard cubic meters, Sm
3
, where the standard conditions are
15°C, 1 bar. In the U.S. gas industry, prefix M represents 10
3
, and MM, B, and T represent 10
6
, 10
9
,
and 10
12
, respectively. We use this convention for both engineering and SI units.
advantages over petroleum and coal, as shown in Table 1.1. Carbon dioxide, a
© 2006 by Taylor and Francis Group, LLC
2 Fundamentals of Natural Gas Processing
when oil or coal is burned. Particulate formation is significantly less in gas
compared with coal and oil, an important environmental consideration because
in addition to degrading air quality, high levels of particulates may pose significant
health problems.
The values reported in Table 1.1 for sulfur dioxide can be misleading. Many
natural gases contain considerable quantities of sulfur at the wellhead, but specifi-
cations for pipeline-quality gas require almost total sulfur removal before pipelining
and sale. Consequently, the tabular values for natural gas represent combustion after
removal of sulfur compounds, whereas the tabular values for oil and coal are
reported for fuels with no sulfur recovery either before or after combustion. Nev-
ertheless, gas produces far fewer pollutants than its competitors, and demand for
gas, the clean fuel, is expected to rise significantly in the near future.
1.1.1 WORLD PICTURE FOR NATURAL GAS
dry natural gas (natural gas with natural gas liquids [NGLs] removed) is on a par
with coal in importance.
almost half of the reserves located in Iran and Russia. The total reported natural
gas reserves (~6,040 Tcf [171 TSm
3
] at the beginning of 2005 [Energy Informa-
tion Administration, 2005c]) do not include discovered reserves that are not
economically feasible to bring to market. This “stranded gas” resides in remote
regions, where the reserve size does not justify the cost of the infrastructure
required to bring it to market. Note that proven reserve estimates are truly
TABLE 1.1
Pounds of Air Pollutants Produced per Billion Btu of Energy
Pollutant Natural Gas
a
Oil
b
Coal
c
Carbon dioxide 117,000 164,000 208,000
Carbon monoxide 40 33 208
Nitrogen oxides 92 448 457
Sulfur dioxide 0.6 1,122 2,591
Particulates 7.0 84 2,744
Formaldehyde 0.750 0.220 0.221
Mercury 0.000 0.007 0.016
a
Natural gas burned in uncontrolled residential gas burners.
b
Oil is # 6 fuel oil at 6.287 million Btu per barrel and 1.03% sulfur with no
postcombustion removal of pollutants.
c
Bituminous coal at 12,027 Btu per pound and 1.64% sulfur with no postcombustion
removal of pollutants.
Source: Energy Information Administration (1998).
The current status of primary energy sources is summarized in Figure 1.1. Basically,
Six countries possess two thirds of the world’s gas reserves (Figure 1.2), with
© 2006 by Taylor and Francis Group, LLC
Overview of the Natural Gas Industry 3
FIGURE 1.1 Primary sources of energy in the world in 2003. Total energy used was 405
quadrillion Btu (Energy Information Administration, 2005b).
FIGURE 1.2 Major proven natural gas reserves by country. Total proven reserves esti-
mated to be 6,040 Tcf (Energy Information Administration, 2005c).
Coal, 24.1%
Natural gas,
23.5%
Crude oil,
35.3%
NGL, 2.6%
Nuclear, 6.6%
Hydroelectric, 6.6%
All other, 1.4%
0
200
400
600
800
1,000
1,200
1,400
1,600
1,800
R
u
s
s
i
a
I
r
a
n
Q
a
t
a
r
S
a
u
d
i

A
r
a
b
i
a
U
n
i
t
e
d

A
r
a
b
E
m
i
r
a
t
e
s
U
n
i
t
e
d
S
t
a
t
e
s

N
i
g
e
r
i
a
A
l
g
e
r
i
a
V
e
n
e
z
u
e
l
a
I
r
a
q
P
r
o
v
e
n

r
e
s
e
r
v
e
s
,

T
c
f

© 2006 by Taylor and Francis Group, LLC
4 Fundamentals of Natural Gas Processing
estimates and vary among sources. Also, proven reserves depend on gas prices;
increased gas price causes reserve estimates to rise.
The world production of natural gas is summarized in Table 1.2. Noteworthy are
the relationships between production and reserves in North America and Eastern
Europe and the high percentage of gas flared or vented in Africa. North America
(principally the United States) has the world’s second largest production of dry gas
and accounts for 29% of world production but possesses only 5% of the reserves.
Eastern Europe slightly leads North America in dry gas production but has 36% of
TABLE 1.2
World Natural Gas Production and Estimated Proven
Reserves at End of 2002
Region
Gross
Production
a
Vented
or
Flared
a
Reinjected
a

Marketed
Production
a
Dry
Gas
Production
a
Proven
Reserves
b
North
America
33,060
(936)
29.5%
176
(4.98)
6.3%
3.895
(110)
31.0%
28,487
(807)
29.5%
26,893
(762)
29.2%
255,800
(7,243)
4.6%
Central and
South
America
5,983
(169)
5.3%
350
(9.91)
12.5%
1,404
(39.76)
11.2%
4,229
(120)
4.4%
3,722
(105)
4.0%
250,100
(7,082)
4.5%
Western Europe 12,333
(349)
11.0%
135
(3.82)
4.8%
1,236
(35.0)
9.8%
10,963
(310)
11.4%
10,548
(299)
11.4%
191,600
(5,426)
3.5%
Eastern Europe
and former
U.S.S.R.
27,047
(766)
24.1%
253
c
(7.16)
9.1%
1
(0.03)
0.0%
27,046
(766)
28.0%
27,046
(766)
29.3%
1,964,200
(55,620)
35.7%
Middle East 12,667
(359)
11.3%
413
(11.69)
14.8%
2,696
(76.34)
21.4%
9,558
(271)
9.9%
8,674
(246)
9.4%
1,579,700
(44,732)
28.7%
Africa 9,450
(268)
8.4%
1,241
(35.14)
44.5%
3,007
(85.15)
23.9%
5,202
(147)
5.4%
4,741
(134)
5.1%
418,200
(11,842)
7.6%
Asia and
Oceania
11,637
(330)
10.4%
224
(6.34)
8.0%
331
(9.37)
2.6%
11,083
(314)
11.5%
10,528
(298)
11.4%
445,400
(12,612)
8.1%
World total 112,178
(3,177)
2,792
(79.06)
12,570
(355.94)
96,568
(2,735)
92,152
(2,609)
5,504,900
(155,881)
a
Data from Energy Information Administration (2005d).
b
Data from Energy Information Administration (2004a).
c
Value given is for 1998 as an estimate because value for 2002 was unreported.
Values are in Bcf (BSm
3
) and percentage values are percent of world total.
© 2006 by Taylor and Francis Group, LLC
Overview of the Natural Gas Industry 5
the world reserves; three quarters of those reserves are located in Russia. Africa vents
or flares 13% of gross production, an exceptionally high number considering that the
world average, excluding Africa, is an estimated 2.3%. The disproportionately high
loss in Africa is caused by the lack of infrastructure in many of the developing nations.
Nigeria alone flares 2 MMscfd (56 MSm
3
/d*), which is equivalent to the total annual
power generation in sub-Saharan Africa. An effort is underway to reduce flaring and
to convert much of the gas to LNG for export (Anonymous, 1999).
1.1.2 NATURAL GAS IN UNITED STATES
Natural gas plays an extremely important role in the United States and accounts for
approximately 23% of the total energy used. Figure 1.3 shows the relationship among
energy sources in the United States, as well as projected growth through 2025. Gas
is presently second only to petroleum, and the difference in demand for gas over
coal is expected to increase substantially with time. Of interest is the prediction that
energy from nuclear and hydroelectric sources will be flat, and nonhydroelectric
renewables are not expected to play a significant role through 2025.
The distribution of natural gas from the wellhead through consumption is shown
of the gross gas produced (14%) are returned to the reservoir for repressurization
of the field. Second, the loss of gas because of venting or flaring is quite small,
FIGURE 1.3 United States energy consumption by fuel. (Adapted from Energy Informa-
tion Administration, 2005a.)
* In this book the symbol M represents 1000 for both engineering and SI units.
0
10
20
30
40
50
60
70
1970 1980 1990 2000 2010 2020 2030
Year
E
n
e
r
g
y

c
o
n
s
u
m
p
t
i
o
n
,

q
u
a
d

B
T
U
Actual consumption Projected consumption
Petroleum
Natural gas
Coal
Nuclear
Nonhydro renewables
Hydro
in Figure 1.4. The numbers reveal some significant points. First, substantial amounts
© 2006 by Taylor and Francis Group, LLC
6 Fundamentals of Natural Gas Processing
only 0.4% of the gross withdrawal. Third, the nonhydrocarbon gases removed (2.5%
of gross) occur in sufficient quantities to render the gas unmarketable, and the
extraction losses (4.1% of gross) refer to liquids (NGL) removed from the gas and
sold separately. Fourth, the imports that account for approximately 18% of the
consumption come predominately from Canada.
In November, 2005, the average wellhead, city gate, and residential prices
were $9.84, $11.45 and $15.80 per thousand cubic feet, respectively (Energy
Information Agency, 2006 i).
for approximately 19% of consumption, but, of that amount, LNG imports are
only 2.9% of total consumption. Also worthy of note is that proven reserves in
2004 constituted only an 8-year supply at the current rate of consumption.
FIGURE 1.4 Natural gas supply and disposition in the United States in 2003. Values
shown are in Tcf. (Adapted from Energy Information Administration, 2005d.)
Dry gas
production
19.0
Nonhydrocarbon gases
removed
0.5
Gross withdrawals
from gas and oil wells
24.1
Vented/flared
0.1
Canada
3.490
Extraction loss
0.9
Reservoir
repressuring
3.5
Imports
Trinidad
0.378
Algeria
0.053
Nigeria
0.050
Qatar
0.014
Oman
0.009
Malaysia
0.003
Canada
0.294
Exports
Mexico
0.064
Japan
0.064
Additions
3.360
Withdrawals
3.2
Natural gas
storage facilities
Gas industry use
1.8
Residential
5.078
Commercial
3.217
Vehicle fuel
0.018
Industrial
7.139
Electric power
5.135
Table 1.3 shows that in the area of production and reserves, imports account
© 2006 by Taylor and Francis Group, LLC
Overview of the Natural Gas Industry 7
1.1.3 NONCONVENTIONAL GAS RESERVES IN UNITED STATES
At present, the two major potential nonconventional gas sources are coal bed
methane (CBM) and naturally occurring gas hydrates. The United States Geo-
logical Survey (USGS) estimates 700 Tcf (20 TSm
3
) of CBM in the United States,
but only 100 Tcf (3 TSm
3
) are recoverable with existing technology (Nuccio,
2000). The most active region is the Powder River Basin area of Wyoming and
Montana. Environmental concerns may limit production (National Petroleum
Technology Office, 2004).
and in sediments of permafrost regions, such as northern Canada and Alaska. The
USGS estimates about 320,000 Tcf (9,000 TSm
3
) of methane in hydrates in the
United States; one half of that reserve is in offshore Alaska (Collett, 2001). An
estimated 45 Tcf (1.2 TSm
3
) in gas hydrates is on the North Slope of Alaska,
where oil is currently produced. These reserves would be the most economically
attractive to produce because the hydrates are concentrated, and much of the
infrastructure for gas processing already exists. However, for the gas to reach the
market, a pipeline must be built.
1.2 SOURCES OF NATURAL GAS
Conventional natural gas generally occurs in deep reservoirs, either associated with
crude oil (associated gas) or in reservoirs that contain little or no crude oil (nonasso-
ciated gas). Associated gas is produced with the oil and separated at the casinghead
TABLE 1.3
Natural Gas in the United States, 2004
U.S. Production 18 Tcf (510 BSm
3
)
U.S. Consumption 22.4 Tcf (634 BSm
3
)
U.S. Imports 4.2 Tcf (120 BSm
3
)
U.S. Exports 0.85 Tcf (24 BSm
3
)
Wellhead price $5.49/Mscf
Average city-gate price
a
$6.65/Mscf
Average price to residential customers $10.74/Mscf
Average price to commercial customers $9.26/Mscf
Average price to industrial customers $6.41/Mscf
Average price to electrical utilities $5.56/Mscf
LNG imports 0.65 Tcf (18 BSm
3
)
Number of producing gas and gas condensate wells (2003) 393,327
Pipeline miles (2003) 306,000 (492,000 km)
Pipeline capacity (2003) 178 Bcfd (5 BSm
3
/day)
Dry natural gas proven reserves 189 Tcf (5.35 BSm
3
)
a
City gate is the point where the gas is transferred from the pipeline to the distribution facilities.
Source: Energy Information Administration (2005g).
Naturally occurring gas hydrates (see Chapter 3) form on the ocean bottom
© 2006 by Taylor and Francis Group, LLC
8 Fundamentals of Natural Gas Processing
or wellhead. Gas produced in this fashion is also referred to as casinghead gas, oil
well gas, or dissolved gas. Nonassociated gas is sometimes referred to as gas-well
gas or dry gas. However, this dry gas can still contain significant amounts of NGL
components. Roughly 93% of the gas produced in the United States is nonassociated
(Energy Information Administration, 2004b). A class of reservoirs, referred to as gas
condensate reservoirs, occurs where, because of the high pressures and temperatures,
the material is present not as a liquid or a gas but as a very dense, high-pressure fluid.
Figure 1.5 shows a simplified flow of material from reservoir to finished
product and provides an overall perspective of the steps involved in taking natural
FIGURE 1.5 Schematic overview of natural gas industry. (Adapted from Cannon, 1993.)
Oil
Water
Gas
Water
Gas
Oil
well
Gas
well
Gas
well
Lease
separator
Field treating Compression
Ethane
Propane
i-Butane
i-Butane
n-Butane
Natural gasoline
Condensate
Natural gas
Gas processing
plant
Treating
systems
Fractionation
systems
Raw natural gas
Crude
oil
G
a
s

l
i
f
t

&
r
e
i
n
j
e
c
t
i
o
n

C
O
2

f
o
r

E
O
R
Casinghead gas
Lease
separator
© 2006 by Taylor and Francis Group, LLC
Overview of the Natural Gas Industry 9
gas from the wellhead to the customer. The chapters that follow provide more
systems. These systems typically are complex, and they bring gas from many
fields and leases to gas plants.
Some gas plants receive feeds from refineries. These streams differ from
natural gases in that they can contain propylene and butylene. They may also
contain trace amounts of undesirable nitrogen compounds and fluorides. This
book considers only the processing of gas and liquids coming directly from gas
and oil leases.
1.3 NATURAL GAS COMPOSITIONS
1.3.1 TRADITIONAL NATURAL GAS
Traditional natural gases, that is, associated and unassociated gas from wells,
vary substantially in composition. Table 1.4 shows a few typical gases. Water is
almost always present at wellhead conditions but is typically not shown in the
analysis. Some gas fields, however, contain no water. Unless the gas has been
dehydrated before it reaches the gas processing plant, the common practice is to
assume the entering gas is saturated with water at the plant inlet conditions.
TABLE 1.4
Typical Gas Compositions
Canada
(Alberta)
Western
Colorado
Southwest
Kansas
Bach Ho
Field
a

Vietnam
Miskar
Field
Tunisia
Rio Arriba
County,
New Mexico
Cliffside
Field,
Amarillo,
Texas
Helium 0.0 0.0 0.45 0.00 0.00 0.0 1.8
Nitrogen 3.2 26.10 14.65 0.21 16.903 0.68 25.6
Carbon
dioxide
1.7 42.66 0.0 0.06 13.588 0.82 0.0
Hydrogen
sulfide
3.3 0.0 0.0 0.00 0.092 0.0 0.0
Methane 77.1 29.98 72.89 70.85 63.901 96.91 65.8
Ethane 6.6 0.55 6.27 13.41 3.349 1.33 3.8
Propane 3.1 0.28 3.74 7.5 0.960 0.19 1.7
Butanes 2.0 0.21 1.38 4.02 0.544 0.05 0.8
Pentanes
and
heavier
3.0 0.25 0.62 2.64 0.630 0.02 0.5
a
Tabular mol% data is on a wet basis (1.3 mol% water)
Source: U.S. Bureau of Mines (1972) and Jones et al. (1999).
detail on the various steps. Note that Figure 1.5 oversimplifies the gas gathering
© 2006 by Taylor and Francis Group, LLC
10 Fundamentals of Natural Gas Processing
1.3.2 IMPORTANT IMPURITIES
A number of impurities can affect how the natural gas is processed:
Water. Most gas produced contains water, which must be removed. Con-
centrations range from trace amounts to saturation.
Sulfur species. If the hydrogen sulfide (H
2
S) concentration is greater than
2 to 3%, carbonyl sulfide (COS), carbon disulfide (CS
2
), elemental sulfur,
and mercaptans* may be present.
Mercury. Trace quantities of mercury may be present in some gases; levels
reported vary from 0.01 to 180 µg/Nm
3
. Because mercury can damage
the brazed aluminum heat exchangers used in cryogenic applications,
conservative design requires mercury removal to a level of 0.01 µg/Nm
3
(Traconis et al., 1996)
NORM. Naturally occurring radioactive materials (NORM) may also
present problems in gas processing. The radioactive gas radon can occur
in wellhead gas at levels from 1 to 1,450 pCi/l (Gray, 1990).
have extreme amounts of undesirable components. For example, accord-
ing to Hobson and Tiratso (1985), wells that contain as much as 92%
carbon dioxide (Colorado), 88% hydrogen sulfide (Alberta, Canada),
and 86% nitrogen (Texas) have been observed.
Oxygen. Some gas-gathering systems in the United States operate below
atmospheric pressure. As a result of leaking pipelines, open valves, and
other system compromises, oxygen is an important impurity to monitor.
A significant amount of corrosion in gas processing is related to oxygen
ingress.
1.3.3 COAL BED METHANE
Coal beds contain large amounts of natural gas (usually referred to as coal bed
methane, or CBM) that is adsorbed on the internal surfaces of the coal or absorbed
within the coal’s molecular structure. This gas can be produced in significant
quantities from wells drilled into the coal seam by lowering the reservoir pressure.
As is the case with conventional natural gas, the composition of the CBM
produced varies widely. In addition to methane, these gases may contain as much
as 20% ethane and heavier hydrocarbons, as well as substantial levels of carbon
dioxide. However, a typical CBM analysis would reveal water saturation, up to
10% carbon dioxide, up to 1% nitrogen, no or very small amounts of ethane and
heavier hydrocarbons, and a balance of methane. Because water is normally
* Mercaptans are highly reactive and odiferous, organic compounds with the formula RSH, in which
R represents an alkane group. Natural gases typically contain methyl through amyl mercaptans. The
ethyl and propyl mercaptans are added to natural gas and propane as odorants. They received their
name from being reactive with mercury. The compounds readily oxidize in the presence of air and
metal to form disulfides that are nearly odorless.
Diluents. Although the gases shown in Table 1.4 are typical, some gases
© 2006 by Taylor and Francis Group, LLC
Overview of the Natural Gas Industry 11
present in the reservoir, it is produced in significant amounts along with the CBM,
and this produced water can pose a significant problem because it may contain
large quantities of dissolved solids that make it unfit for domestic or agricultural
uses (National Petroleum Technology Office, 2004).
1.3.4 SUBQUALITY GAS
The Gas Research Institute (Meyer, 2000) classified natural gases from the lower
48 states as high quality and subquality. Subquality is divided into seven catego-
ries, depending on the amount of N
2
, CO
2
, and H
2
S present. For their definition
of subquality. The gas contains more than 2% CO
2
, 4%N
2
, and 4 ppmv H
2
S.
Table 1.5 summarizes the evaluation for proven raw reserves.
1.4 CLASSIFICATION
Natural gases commonly are classified according to their liquids content as either
lean or rich and according to the sulfur content as either sweet or sour. This
section provides some quantification of these qualitative terms.
1.4.1 LIQUIDS CONTENT
Gas composition plays a critical role in the economics of gas processing. The more
liquids, usually defined as C
2
+, in the gas, the “richer” the gas. Extraction of these
liquids produces a product that may have a higher sales value than does natural gas.
To quantify the liquids content of a natural gas mixture, the industry uses
GPM, or gallons of liquids recoverable per 1,000 standard cubic feet (Mscf) of
gas. (In metric units, the quantity is commonly stated as m
3
of liquid per 100 m
3
TABLE 1.5
Quality of Proven Natural Gas Reserves of the Lower 48
United States in 1998
Bcf (BSm
3
) %
High quality 87,679 (2,464) 59
Subquality 60,698 (1,699) 41
High N
2
only 15,617 (424) 11
High CO
2
only 17,932 (481) 12
High H
2
S only 5 691 (161) 4
High N
2
& CO
2
1,577 (29) 1
High N
2
& H
2
S 600 (17) 0
High CO
2
& H
2
S 12,697 (340) 9
High N
2
, CO
2
, & H
2
S 6,585 (170) 4
Total 148,377 (4,191) 100
Source: Meyer (2000).
© 2006 by Taylor and Francis Group, LLC
12 Fundamentals of Natural Gas Processing
of gas.) The term usually applies to ethane and heavier components but sometimes
applies instead to propane and heavier components. Determination of the GPM
requires knowledge of the gas composition on a mole basis and the gallons of
higher hydrocarbons. Note that ethane is not a liquid at 60°F (15.5°C), so the
value is a hypothetical value accepted throughout the industry. Also, the actual
volume of liquid obtained from a gas will be less than the GPM value because
complete recovery of ethane and propane is impractical for two reasons:
1. Cost. The low temperature and high compression energy required generally
makes recovery of more than about 90 to 95% of the ethane, 98% of the
propane, and 99% of the butanes uneconomical. Higher ethane recovery
plants also have higher recovery of propane and heavier components.
2. Heating value specifications. As discussed below, a specification applies
to the heating value of gas. Unless the gas contains no nonflammable
diluents (i.e., N
2
and CO
2
), additional hydrocarbons must be in the gas
to obtain the required heating value.
Computation of the GPM requires summation of the product of the number of moles
of each component in 1,000 scf of gas by the gallons of liquid per mole for that
component.
Basis: 1,000 scf of gas.
A lb-mole of gas at standard conditions has a volume of 379.49 ft
3
. This volume
translates into 1,000/379.49, or 2.6351 lb-moles for 1,000 scf. This value is multi-
plied by the mole fraction of each component in the gas and by the gallons of liquid
for each component. Table 1.6 summarizes the calculations.
TABLE 1.6
Calculation of GPM of Alberta Gas
Mole % Moles Gal/Mole GPM
Helium 0 0.0000 0 0
Nitrogen 3.2 0.0843 0 0
Carbon dioxide 1.7 0.0448 0 0
Hydrogen sulfide 3.3 0.0870 0 0
Methane 77.1 2.0317 0 0
Ethane 6.6 0.1739 10.126 1.761
Propane 3.1 0.0817 10.433 0.852
Butanes 2.0 0.0527 12.162 0.641
Pentanes and heavier 3.0 0.0791 13.713 1.084
Totals 100.0 2.6351 4.338
Example 1.1 Calculate the GPM of the Alberta gas given in Table 1.4.
liquid per lb-mole. See Appendix B for the gallons per lb-mole for ethane and
© 2006 by Taylor and Francis Group, LLC
Overview of the Natural Gas Industry 13
For this example, the Gal/mole for butanes was taken as the average of isobutane
and n-butane; the value for C
5
+ was taken to be that of pure n-pentane. The resulting
GPM for this gas is 4.34.
The rich and lean terms refer to the amount of recoverable hydrocarbons present.
The terms are relative, but a lean gas will usually be 1 GPM, whereas a rich gas
may contain 3 or more GPM. Thus, the gas described above is considered fairly rich.
1.4.2 SULFUR CONTENT
Sweet and sour refer to the sulfur (generally H
2
S) content. A sweet gas
contains negligible amounts of H
2
S, whereas a sour gas has unacceptable
quantities of H
2
S, which is both odiferous and corrosive. When present with
water, H
2
S is corrosive. The corrosion products are iron sulfides, FeS
X
, a fine
black powder. Again, the terms are relative, but generally, sweet means the
gas contains less than 4 ppmv of H
2
S. The amount of H
2
S allowable in
pipeline-quality gas is between 0.25 and 1.0 grains per 100 scf (6 to 24
mg/Sm
3
, 4 to 16 ppmv).
1.5 PROCESSING AND PRINCIPAL PRODUCTS
The two primary uses for natural gas are as a fuel and as a petrochemical
feedstock, and consequently, the three basic reasons for processing raw natural
gas are the following:
• Purification. Removal of materials, valuable or not, that inhibit the use
of the gas as an industrial or residential fuel
• Separation. Splitting out of components that have greater value as
petrochemical feedstocks, stand alone fuels (e.g., propane), or indus-
trial gases (e.g., ethane, helium)
• Liquefaction. Increase of the energy density of the gas for storage or
transportation
Depending on the situation, a process may be classified as either separation
or purification. For example, if a small amount of H
2
S is removed, incinerated,
and vented to the atmosphere, the process is purification, but if large amounts of
H
2
S are removed and converted to elemental sulfur, often a low-priced commod-
materials present in natural gas and the slate of possible products from the gas
plant.
Although the principal use of natural gas is the production of pipeline-
quality gas for distribution to residential and industrial consumers for fuel, a
number of components in natural gas are often separated from the bulk gas and
sold separately.
ity, the process is considered separation. Figure 1.6 provides an overview of the
© 2006 by Taylor and Francis Group, LLC
14 Fundamentals of Natural Gas Processing
1.5.1 METHANE
The principal use of methane is as a fuel; it is the dominant constituent of pipeline
quality natural gas. Considerable quantities of methane are used as feedstock in
the production of industrial chemicals, principally ammonia and methanol.
1.5.2 ETHANE
The majority of the ethane used in the United States comes from gas plants, and
refineries and imports account for the remainder. In addition to being left in the
gas for use as a fuel, ethane is used for the production of ethylene, the feedstock
for polyethylene.
1.5.3 PROPANE
Gas plants produce about 45% of the propane used in the United States, refineries
contribute about 44%, and imports account for the remainder. The principal uses
are petrochemical (47%), residential (39%), farm (8%), industrial (4%), and
transportation (2%) (Florida Propane Gas Council, 2005). A special grade of
propane, called HD-5, is sold as fuel.
1.5.4 ETHANE–PROPANE MIX
When NGL is fractionated into various hydrocarbon streams, the butanes along
with part of the propane are sometimes separated for use in local markets because
FIGURE 1.6 Generic raw gas and product slate.
Gas
processing
Raw gas
Water
Helium
Nitrogen
Carbon dioxide
Hydrogen sulfide
Methane
Ethane
Propane
n-Butane
i-Butane
Pentanes +
Water
Nitrogen
Carbon dioxide
Hydrogen sulfide
Pipeline gas
Ethane
Propane
n-Butane
i-Butane
Natural gasoline
Helium
Product Slate
© 2006 by Taylor and Francis Group, LLC
Overview of the Natural Gas Industry 15
they are transportable by truck. The remaining light ends, an ethane−propane mix
(E-P mix), is then pipelined to a customer as a chemical or refining feedstock.
1.5.5 ISOBUTANE
Approximately 42% of the United States supply of isobutene comes from gas
plants, refineries supply about 5% (this percentage does not include consumption
of isobutane within the refinery), and imports are responsible for about 12%. The
remaining isobutane on the market is furnished by isomerization plants that
convert n-butane to isobutane. The three primary markets for isobutane are as a
feedstock for MTBE (methyl tertiary butyl ether) production (which is being
phased out), as a feedstock in the production of reformulated gasoline, and as a
feedstock for the production of propylene oxide.
1.5.6 n-BUTANE
Gas plant production of n-butane accounts for about 63% of the total supply,
refineries contribute approximately 31%, and imports account for the remainder.
Domestic usage of n-butane is predominantly in gasoline, either as a blending
component or through isomerization to isobutane. Specially produced mixtures
of butanes and propane have replaced halocarbons as the preferred propellant in
aerosols.
1.5.7 NATURAL GAS LIQUIDS
Natural gas liquids (NGL) include all hydrocarbons liquefied in the field or in
processing plants, including ethane, propane, butanes, and natural gasoline. Such
mixtures generated in gas plants are usually referred to as “Y-grade” or “raw product.”
1.5.8 NATURAL GASOLINE
Natural gasoline, a mixture of hydrocarbons that consist mostly of pentanes and
heavier hydrocarbons and meet GPA product specifications, should not be con-
fused with natural gas liquids (NGL), a term used to designate all hydrocarbon
liquids produced in field facilities or in gas plants.
The major uses of natural gasoline are in refineries, for direct blending into
gasoline and as a feedstock for C
5
/C
6
isomerization. It is used in the petrochemical
industry for ethylene production.
1.5.9 SULFUR
Current sulfur production in the United States is approximately 15,000 metric
tons per day (15 MMkg/d); about 85% comes from gas processing plants that
convert H
2
S to elemental sulfur. Some major uses of sulfur include rubber vul-
canization, production of sulfuric acid, and manufacture of black gunpowder
(Georgia Gulf Sulfur Corporation, 2005).
© 2006 by Taylor and Francis Group, LLC
16 Fundamentals of Natural Gas Processing
1.6 PRODUCT SPECIFICATIONS
1.6.1 NATURAL GAS
The composition of natural gas varies considerably from location to location, and
as with petroleum products in general, the specifications for salable products from
gas processing are generally in terms of both composition and performance
criteria. For natural gas these criteria include Wobbe number, heating value, total
inerts, water, oxygen, and sulfur content. The first two criteria relate to combustion
characteristics. The latter three provide protection from pipeline plugging and
corrosion.
Specifications have historically been established in contract negotiations and
no firm, accepted standards exist for all products. Consequently, specifications
for pipeline quality gas listed in Table 1.7 are typical but not definitive.
TABLE 1.7
Specifications for Pipeline Quality Gas
Major Components Minimum Mol% Maximum Mol%
Methane 75 None
Ethane None 10
Propane None 5
Butanes None 2
Pentanes and heavier None 0.5
Nitrogen and other inerts None 3
Carbon dioxide None 2−3
Total diluent gases None 4−5
Trace components
Hydrogen sulfide 0.25−0.3 g/100 scf
(6−7 mg/m
3
)
Total sulfur 5−20 g/100 scf
(115−460 mg/m
3
)
Water vapor 4.0−7.0 lb/MM scf
(60−110 mg/m
3
)
Oxygen 1.0%
Other characteristics
Heating value
(gross, saturated)
950−1,150 Btu/scf
( 35,400−42,800 kJ/m
3
)
Liquids Free of liquid water and hydrocarbons
at delivery temperature and pressure
Solids Free of particulates in amounts deleterious
to transmission and utilization equipment
Source: Engineering Data Book (2004).
© 2006 by Taylor and Francis Group, LLC
Overview of the Natural Gas Industry 17
Hydrocarbon dew point is becoming an issue in some situations. The problem
arises from trace condensation in pipelines, which can cause metering problems.
1.6.2 LIQUID PRODUCTS
As with gases, specifications for liquid products are based upon both composition
and performance criteria. For liquid products, the performance specifications
include Reid vapor pressure, water, oxygen, H
2
S, and total sulfur content. Safety
considerations make vapor pressure especially important for the liquid products
because of regulations for shipping and storage containers. Table 1.8 gives major
presents upper limits of common contaminants, but actual specifications vary,
TABLE 1.8
Major Components and Vapor Pressures of Common Liquid Products
Liquid Product Composition
a
Vapor Pressure
b
at 100°F,
psig, max(at 37.8°C, kPa, max)
High-ethane raw
streams
C
1
C
2
–C
5

1–5 wt%
balance

Ethane–propane
mixes
C
1
C
2

C
3
C
4
+
0.6–1 wt%
22.5–40.5
59.5–77.5
0.2–4.5

High-purity
ethane
C
1
C
2
C
3
C
4
+
1.5–2.5 wt%
90–96
6–5
0.5–3.0

Commercial
propane
Predominantly
C
3
and C
3
=
208
(1,434)
Commercial
butane
Predominantly
C
4
and C
4
=
70
(483)
Commercial
butane–propane
mixes
Predominantly
C
4
and C
3
208
(1,434)
Propane HD-5 C
3
C
3
=
C
4
+
> 90 liq vol%
< 5
< 2.5
208
(1,434)
a
Throughout the book C
1
, C
2
etc, refer to methane, ethane, etc. The “=” denotes an olefin.
The term C
4
+ denotes propane and heavier compounds.
b
Vapor pressure as defined by D1267-02 Standard Test Method for Gage Vapor Pressure of
Liquefied Petroleum (LP) Gases (LP-Gas Method).
Source: Engineering Data Book (2004).
component and vapor-pressure specifications for common liquid products. Table 1.9
© 2006 by Taylor and Francis Group, LLC
18 Fundamentals of Natural Gas Processing
depending upon contractual agreement. Water content specifications are less
stringent for propane and butane because liquid pressures are lower, and hydrate
formation is not such a threat. However, as Table 1.9 indicates, the water level
in some propane products must pass a dryness test, which ensures that the water
content is sufficiently low (< 25 ppmw) to avoid hydrate formation when water
products and others are available in GPA standards.
1.7 COMBUSTION CHARACTERISTICS
1.7.1 HEATING VALUE
One of the principal uses of natural gas is as a fuel, and consequently, pipeline
gas is normally bought and sold (custody transfer) on the basis of its heating
value. Procedures for calculating the heat effect in any chemical reaction are
found in standard texts on thermodynamics (e.g., Smith et al., 2001).
Determination of the heating value of a fuel involves two arbitrary but con-
ventional standard states for the water formed in the reaction:
1. All the water formed is a liquid (gross heating value, frequently called
higher heating value [HHV])
2. All the water formed is a gas (net heating value, frequently called lower
heating value [LHV])
TABLE 1.9
Maximum Levels of Major Contaminants of Common Liquefied Products
Concentrations are in ppmw unless specified otherwise.
H
2
S Total Sulfur
a
CO
2
O
2
H
2
O
High-ethane raw streams 50 200 3,500 No free
b
Ethane−propane mixes #1
a
143 3,000 1,000 No free
b
High–Purity ethane 10 70 5,000 5 No free
b
Commercial propane #1 185 — — Pass test
c
Commercial butane #1 140 — — —
Commercial butane−propane mixes #1 140 — — —
Propane HD-5 #1 123 — — Pass test
c
a
Concentration acceptable provided the copper strip test, which detects all corrosive compounds, is
passed. The #1 represents the passing score on the copper-strip test, D1838-05 Standard Test Method
for Copper Strip Corrosion by Liquefied Petroleum (LP) Gases. Eckersley and Kane (2004) discuss
sample handling problems related to the test.
b
Limit is no free water present in product.
c
Moisture level must be sufficiently low to pass the D2713-91(2001) Standard Test Method for
Dryness of Propane (valve freeze method), which corresponds to roughly 10 ppmw.
Source: Engineering Data Book (2004).
is vaporized through an orifice (see Chapter 3). Complete specifications for these
© 2006 by Taylor and Francis Group, LLC
Overview of the Natural Gas Industry 19
The gas industry always uses the gross heating value in custody transfer.
Obviously, the numerical difference between the two heating values is the heat
of condensation of water at the specified conditions. Both states are hypothetical
because the heating value is normally calculated at 60°F and 1 atm (15.6°C and
1.01 atm), standard conditions for the gas industry, and, thus at equilibrium, the
water would be partially liquid and partially vapor. A common practice is also
to assume ideal gas behavior, and consequently, the heating values commonly
calculated and reported are representative of, but not identical to, the values
obtained when the fuel is burned in an industrial or residential furnace.
Heating values for custody transfer are determined either by direct measure-
ment, in which calorimetry is used, or by computation of the value on the basis
of gas analysis. The method is set in the sales contract. The formulas for the
calculation of ideal gas gross heating values, on a volumetric basis are (Gas
Processors Association, 1996)
(1.1)
(1.2)
The equations assume that the gas analysis is given on a dry basis, that
water is x
W
when the gas is saturated at the specified conditions. The mole fraction
can be calculated from
(1.3)
The vapor pressure of water at 60°F (15.6°C), the common base temperature,
is 0.25636 psia (1.76754 kPa). The most commonly used base pressures, P
b
, and
the values of (1− x
W
) are listed below.
The situation regarding water is further complicated by the fact that gas analyses
are normally given on a dry basis, even though the gas may be partially or fully
saturated with water. Consequently, heating value may be calculated on a dry basis,
wet (saturated) basis, or, if the humidity is known, a partially saturated basis.
P
b
(psia) 1 – x
W
14.50 0.9823
14.65 0.9825
14.696 0.9826
14.73 0.9826
15.025 0.9829
H dry x H
v
id
i
i
n
id
vi
( ) =
=

1
H sat x x H
v
id
w i
i
n
vi
id
( ) ( ) = −
=

1
1
H
v
id
x
P
P
w
w
Sat
b
=
is the ideal gross heating value (see Appendix B), and that the mole fraction of
© 2006 by Taylor and Francis Group, LLC
20 Fundamentals of Natural Gas Processing
A complete discussion of heating value calculations, including correction of
the ideal gas values to the real-gas state by use of calculated compressibility factors
(z), is available in GPA Standard 2172-96 (Gas Processors Association, 1996).
Assume the heating value for the butanes to be that of isobutene, and for the C
5
+
fraction, use pure hexane.
Table 1.10 shows the calculations with heating values obtained from Appendix B.
This mixture has a gross heating value of 1,202.2 Btu/scf (44,886 kJ/Sm
3
).
Note that credit is not given for the heating value associated with H
2
S in contractual
situations. It is unlikely that a gas stream with 3.3% H
2
S would be burned.
1.7.2 WOBBE NUMBER
In gas appliances, maintenance of the same combustion characteristics are
desirable when one gas composition is switched to another. Several factors must
be considered, but one of the more important considerations is maintenance of
the same heat release at the burner for a given pressure drop through a control
valve. This combustion characteristic is measured by the Wobbe number,
defined as the gross heating value (Btu/scf) of the gas divided by the square
root of the specific gravity (the ratio of the density of the gas divided by the
density of air; both densities evaluated at the same pressure and temperature).
Two gases with the same Wobbe number are interchangeable as far as heat
release at the burner is concerned.
WB = (gross heating value)/(specific gravity)1/2,
with the specific gravity correcting for flow through an orifice. The Wobbe number
normally has a value between 1,100 and 1,400. The Wobbe number is calculated from
TABLE 1.10
Mole %
Btu/scf
Helium 0 0.0 0
Nitrogen 3.2 0.0 0
Carbon dioxide 1.7 0.0 0
Hydrogen sulfide 3.3 637.1 21.0
Methane 77.1 1010.0 778.7
Ethane 6.6 1769.7 116.8
Propane 3.1 2516.2 78.0
Butanes as isobutane 2.0 3252.0 65.0
Pentanes and heavier as hexane 3.0 4756.0 142.7
Totals 100.0 1202.2
H
vi
id
x H
i vi
id
Example 1.2 Calculate the heating value of the Alberta gas given in Table 1.4.
Calculations with Heating Values Obtained from Appendix B
© 2006 by Taylor and Francis Group, LLC
Overview of the Natural Gas Industry 21
the gross heating value (Btu/scf) and specific gravity of the mixture, not from an
average of the Wobbe numbers of the constituents of the mixture. In Europe a value
of 1400 is generally required.
Some typical Wobbe numbers are
Wobbe numbers are often adjusted by blending the natural gas with air. For
example, one distribution company maintains a Wobbe number between 1,130
and 1,280 for gas distribution to its residential customers by blending air with
the natural gas in three blending stations. A typical set of operating conditions
for one of the air-blending stations is given below.
Because air blending also changes the heating value of the gas, a balance must
be maintained between the Wobbe number and the gross heating value.
The Wobbe number is more commonly used in Europe. However, it will
become more important in the United States in the future as LNG importation
increases. Heating values for LNG can be much higher than typical sales gas
from a gas plant in the United States.
REFERENCES
Anonymous, Harnessing Abundant Gas Reserves, Africa Recovery, United Nations, 13 (1)
August 2005.
BP Statistical Review of World Energy 2005.
Cannon, R.E., The Gas Processing Industry, Origins and Evolution, Gas Processors Asso-
ciation, Tulsa, OK, 1993.
Collett, T., Natural Gas Hydrates: Vast Resource, Uncertain Future, U.S. Geological Survey Fact
Eckersley, N. and Kane, J.A., Designing customized desulfurization systems for the treat-
ment of NGL streams, in the Proceedings of the Laurance Reid Gas Conditioning
Conference, Norman, OK, 2004.
Energy Information Administration, U.S. Department of Energy, Natural Gas 1998, Issues
Methane 1,360 Btu/scf
Ethane 1,740 Btu/scf
Propane 2,044 Btu/scf
80% methane + 20% ethane 1,443 Btu/scf
95% methane + 5% ethane 1,381 Btu/scf
Inlet Wobbe 1,335
Inlet Btu/scf 1,080 (40,324 kJ/Sm
3
)
Inlet specific gravity 0.654
Normal outlet Wobbe 1,210
Peak day outlet Wobbe 1,280
1999, http://www.un.org/ecosocdev/geninfo/afrec/vol13no1/jun99.htm, Retrieved
http://www.bp.com/genericsection.do?categoryId=92&contentId=7005893
Sheet FS-021-01, 2001, http://pubs.usgs.gov/fs/fs021-01/, Retrieved October 2005.
and Trends, 1999, www.eia.doe.gov/oil_gas/natural_ gas/ analysis_publications/
natural_gas_1998_issues_and_trends/it98.html, Retrieved September 2005.
© 2006 by Taylor and Francis Group, LLC
22 Fundamentals of Natural Gas Processing
Energy Information Administration, U.S. Department of Energy, Natural Gas Annual 2003,
Energy Information Administration, U.S. Department of Energy, 2004 United States Total
Energy Information Administration, U.S. Department of Energy, Annual Energy Outlook
2005 Market Trends—Energy Demand 2005, 2002, Figure 52, Industrial Delivered
Energy Consumption By Fuel, 1970−2025,
Energy Information Administration, U.S. Department of Energy, Annual Energy Review,
Energy Information Administration, U.S. Department of Energy, International Energy
Retrieved June 2005.
Energy Information Administration, U.S. Department of Energy, International Energy
Energy Information Administration, U.S. Department of Energy, International Energy Outlook
2004, Figure 40, World Natural Gas Resources by Region As of January 1, 2004, 2005f,
Energy Information Administration, U.S. Department of Energy, Natural Gas Navigator,
2005g, tonto.eia.doe.gov/dnav/ng/ng_sum_lsum_dcu_nus_a.htm, Retrieved June
2005.
Energy Information Administration, U.S. Department of Energy, Changes in U.S. Natural
March, 2006.
Engineering Data Book, Product Specifications, 12th ed., Sec. 2, Gas Processors Supply
Association, Tulsa, OK, 2004.
Retrieved August 2005.
Gas Processors Association, GPA Standard 2172-96, Calculation of Gross Heating Value,
Relative Density, and Compressibility of Natural Gas Mixtures from Composi-
tional Analysis, Tulsa, OK, 1996.
Gray, P., Radioactive materials could pose problems for the gas industry, Oil and Gas J., 88
(26) 45, 1990.
Hobson, G.D., and Tiratso, E.N., Introduction to Petroleum Geology, Gulf Publishing,
Houston, TX, 1985.
Jones, S., Lee, S., Evans, M., and Chen, R., Simultaneous Removal of Water and BTEX
from Feed Gas for a Cryogenic Plant, in Proceedings of the Seventy-Eighth Annual
Convention of the Gas Processors Association, Tulsa, OK, 1999, 108.
Meyer, H.S., Volume and distribution of subquality natural gas in the United States,
GasTIPS, 6, 10, 2000.
2002, Distribution of Wells by Production Rate Bracket, 2004b, http://www.eia.
doe.gov/pub/oil_gas/petrosystem/us_table.html Retrieved June 2005.
Table 11.1, World Primary Energy Production By Source, 1970−2002, 2005b,
2004a. www.eia.doe.gov/pub/oil_gas/natural_gas/data_publications/natural_gas_
annual/current/pdf/nga03.pdf, Retrieved June 2005.
2005a, www.eia.doe.gov/oiaf/
aeo/excel/figure52_data.xls, Retrieved June 2005.
www.eia.doe.gov/aer/txt/ptb1101.html, Retrieved June 2005.
Annual 2003 2005, 2005c, www.eia.doe.gov/emeu/international/reserves.html,
Annual 2003, Table 4.1, World Natural Gas Production, 2002, 2005d, www. eia.
Energy Information Administration, U.S. Department of Energy, International Energy
doe.gov/pub/international/iea2003/table41.xls, Retrieved June 2005.
Annual 2003, Figure 2, 2002, 2005e, www.eia.doe.gov/pub/international/iea 2003
/table41.xls, Retrieved June 2005.
www.eia.doe.gov/oiaf/ieo/excel/figure_40data.xls, Retrieved June 2005.
Energy Information Administration, www.eia.doe.gov/ Retrieved March, 2006.
Florida Propane Gas, Safety, Education and Research Council, www.propanefl.com,
Georgia Gulf Sulfur Corporation, www.georgiagulfsulfur.com, Retrieved August 2005.
Gas Transportation Infrastructure in 2004, 2005h, www.eia.doe.gov/Retrieved
© 2006 by Taylor and Francis Group, LLC
Overview of the Natural Gas Industry 23
National Petroleum Technology Office, U.S. Department of Energy, Coal Bed Methane
Primer, New Source of Natural Gas—Environmental Implications Background
Nuccio, V., Coal-Bed Methane: Potential And Concerns, U. S. Geological Survey Fact Sheet
Smith, J.M., Van Ness, H.C., and Abbott, M.M., Introduction to Chemical Engineering
Thermodynamics, 6th ed., McGraw-Hill, New York, 2001.
Traconis, B., Mierez, Y.D., and Jimenez, A., Mercury Removal Systems at Santa Barbara
Extraction Plant, in Proceedings of the Seventy-Fifth Annual Convention of the
Gas Processors Association, Tulsa, OK, 1996, 123.
U.S. Bureau of Mines, Analyses of Natural Gases, 1972, Information Circular # 8607, 1972.
BIBLIOGRAPHY
For additional information on the natural gas industry and gas processing, the following
books are recommended.
Campbell, J.M., Gas Conditioning and Processing, Volumes 1 and 2. Campbell Petroleum
Series, J.M. Campbell Co., Norman, OK, 1976.
Cannon, R.E., The Gas Processing Industry, Origins and Evolution, Gas Processors Asso-
ciation, Tulsa, OK, 1993.
Ikoku, C.U., Natural Gas Engineering, PennWell Publishing Co., Tulsa, OK, 1980.
Katz, D.L., Cornell, D., Kobayashi, R., Poettmann, F.H., Vary, J.A., Elenbass J.R., and
Weinaug, C. F., Handbook of Natural Gas Engineering, McGraw-Hill, New York,
1959.
Katz, D.L. and. Lee, R.L, Natural Gas Engineering, Production and Storage, McGraw-
Hill, New York, 1990.
Kohl, A., and Nielsen, R., Gas Purification, 5th ed., Gulf Publishing, Houston, TX, 1997.
WEB SITES
tabulation of data, and statistics regarding reserves, production, transportation,
storage, and consumption of energy both in the United States and worldwide.
major industrial organization dealing with all aspects of gas and gas liquids
processing.
Gas Supply Association. It provides a good overview of natural gas from explo-
zation and items of importance to the natural gas industry.
that provides a wide array of data regarding all forms of energy.
programs of the U.S. Department of Energy.
Natural Gas Suppliers Association, History, 2004, http://www.naturalgas.org/overview/
history.asp, Retrieved August 2005.
FS-123-00, 2000, http://pubs.usgs.gov/fs/fs123-00/, Retrieved October 2005.
and Development in the Rocky Mountain West, 2004, www.mines.edu/ research/
PTTC/CBM/primer/CBM%20PRIMER%20FINAL.pdf, Retrieved June 2005.
Energy Information Administration: www.eia.doe.gov. This site provides extensive reports,
Gas Processors Association: www.gasprocessors.org. This site is the home page of the
Gas Industry Consortium: www.naturalgas.org. This site is the home page of the Natural
ration to consumption. The site www.ngsa.org gives an overview of the organi-
International Energy Agency: www.iea.org. This site is the home page of the organization
National Energy Technology Laboratory, Strategic Center for Natural Gas: www.netl.doe.
gov/scngo/. This site is the starting point for learning about energy-related research
© 2006 by Taylor and Francis Group, LLC
25
2
Overview of Gas Plant
Processing
2.1 ROLES OF GAS PLANTS
Gas plants play a variety of roles in the oil and gas industry. The desired end product
dictates the processes required. Some primary purposes of gas plants include:
• Dehydration of gas to reduce corrosion and to prevent gas hydrate
formation. These plants commonly are found on offshore platforms,
where associated gas is separated from oil and dehydrated. Depending
upon pipeline infrastructure, the gas may be recombined with the oil
before it is put into a pipeline to shore.
• Associated oil stabilization. One of the world’s largest gas plants is on
the North Slope of Alaska. The sole purpose of the facility is to strip
associated gas from the oil so that the oil is safe to put into the Aleyska
pipeline. The produced gas is reinjected into the formation to enhance
oil recovery. This gas will be sold if a gas pipeline is built between
the North Slope and the lower 48 states.
• Carbon dioxide or nitrogen recovery for enhanced oil recovery (EOR).
These plants separate natural gas from the CO
2
or N
2
; the natural gas
is marketed and the CO
2
or N
2
are reinjected into formations. In N
2
projects, an air plant may be constructed on site to provide additional
N
2
at the beginning of the project.
• Upgrading subquality gas. To make the gas marketable, the undesired
diluents N
2
, H
2
S, and CO
2
are removed. Of the three components, N
2
is the most difficult to remove because it requires cryogenic processing
when large volumes are processed.
• Helium recovery. Few plants are dedicated primarily to helium recovery.
Therefore, this facility is typically an addition to a gas plant. Natural gas
is the primary source of helium. In the United States, the highest helium
concentrations are in eastern Colorado, southwestern Wyoming, western
Kansas, and the panhandles of Oklahoma and Texas.
• Liquefaction. Some gas plants are dedicated to the production of hydro-
carbon liquids and a natural gas stream to make liquefied natural gas
(LNG). These plants are in locations with large gas reserves and no
pipelines to market. The only large-scale LNG production facility in
the United States is in Alaska. However, the United States has more
than 60 peak shaving plants that produce LNG from pipeline gas for
storage purposes (Energy Information Administration, 2003).
© 2006 by Taylor and Francis Group, LLC
26 Fundamentals of Natural Gas Processing
However, the major role of gas plants in the United States is to process both
associated and nonassociated gas to produce high-quality natural gas and hydro-
carbon liquids. Sale of liquids provides a significant portion of the income from
these plants. Plants optimize profits by adjusting the fraction of liquids recovered
while meeting the specifications for the natural gas.
2.2 PLANT PROCESSES
process elements. An overview of each element is given here and details are
provided in the subsequent chapters. To show how the process elements are inte-
We are unaware of any plants that have all of the process steps shown, and
the less common processes are shaded. Which processes are in a plant is dictated
by the feed gas composition and conditions, along with the desired product
streams.
2.2.1 FIELD OPERATIONS AND INLET RECEIVING
All plants have field operations and a network of pipelines that feed the raw natural
gas and liquids into the plant. Field operations may include dehydration, CO
2
and
H
2
S removal, and compression. These processes are discussed more below. Unless
the gas is completely free of any liquids, once it enters the plant, the gas and liquids
go into inlet receiving, where the initial gas−liquid separation is made. Condensed
water, hydrocarbon liquids, and solids are removed. Water and solids are processed
for disposal, and the hydrocarbon liquids go on to liquids processing, as discussed
also discusses gas hydrates and how to avoid their formation.
2.2.2 INLET COMPRESSION
Most plants have inlet compression, but compression requirements vary. High
pressure is critical, as it drives the cryogenic-liquids recovery process. For inlet
pressures of around 1,000 psi (70 bar) or higher, only gas coming from the liquids
processing step needs compression. However, most onshore gas plants in the
discusses compression. The same types of compressors are used for field and
outlet compression.
2.2.3 GAS TREATING
Most plants have a gas treating step to remove the acid gases H
2
S and CO
2
,

along
with other sulfur impurities. Most plants use water-based absorbents to remove
various methods for removal of these components.
Figure 2.1 is a block schematic of a gas plant that has all of the more common
grated, Chapter 15 discusses several gas plants with different configurations.
below. Chapter 3 discusses both field operations and inlet receiving. The chapter
United States have some low-pressure gas streams entering the plant. Chapter 4
the impurities, but other solvents and processes are used. Chapter 5 discusses the
© 2006 by Taylor and Francis Group, LLC
Overview of Gas Plant Processing 27
2.2.4 DEHYDRATION
Nearly all plants utilize a dehydration step because the gas that leaves the gas
treating step is usually water saturated. Even if no water-based gas treating is
required, most gas streams contain too much water to meet pipeline specifications
or to enter the cryogenic section of the plant. Field operations sometimes dry the
provides details on the commonly used dehydration processes.
FIGURE 2.1 Block schematic of various processes of a gas plant. Processes required
depend upon feed and product slate. Less common process steps for facilities in the United
States are shaded. The heavier lines denote liquid streams.
Field Liquids Removal
Gas & Liquids from Wells
Inlet
Compression
Sulfur
Recovery
Gas Treating
Dehydration
Nitrogen
Rejection
Helium
Recovery
Outlet
Compression
Elemental
Sulfur
Field Dehydration
Field Acid Gas Removal
Field Compression
Sales Gas
Water &
Solids
Natural
Gas Liquids
Natural
Gasoline
Raw Helium
CO
2
Liquefaction
Liquefied
Natural Gas
Inlet Receiving
Hydrocarbon
Recovery
Liquids Processing
gas to avoid gas hydrate formation as well as to reduce corrosion. Chapter 6
© 2006 by Taylor and Francis Group, LLC
28 Fundamentals of Natural Gas Processing
2.2.5 HYDROCARBON RECOVERY
Any plant that processes natural gas to produce hydrocarbon liquids (NGL) or LNG
utilizes a hydrocarbon recovery step. This step usually involves cryogenic separation
to recover the ethane and heavier hydrocarbons. Hydrocarbon recovery often plays
an important role in field operations, where it is used for fuel gas conditioning and
to alter gas condensation temperatures. Details of these processes are provided in
2.2.6 NITROGEN REJECTION
Although a less common process in the gas industry, nitrogen rejection will
become more important as we shift to lower-quality gas feedstock. This process
is typically cryogenic, although membrane and absorbent technology are becom-
2.2.7 HELIUM RECOVERY
Helium recovery is uncommon, unless the helium content is above 0.5 vol%. A
treatment of other trace components, including BTEX (benzene, toluene, ethyl-
benzene, and xylenes) emissions and mercury. BTEX is primarily an environ-
mental concern because of possible emissions from glycol dehydration units.
Although at extremely low concentrations in the gas, elemental mercury can cause
mechanical failure in aluminum heat exchangers.
2.2.8 OUTLET COMPRESSION
Most plants must compress the gas before it goes to the pipeline. The majority
of plants that have cryogenic hydrocarbon recovery use turboexpanders to provide
refrigeration in the cryogenic section. Work generated in expansion is used to
recompress the outlet gas. However, additional compression is usually required.
covers stand-alone compressors.
2.2.9 LIQUIDS PROCESSING
Liquids processing occurs whenever NGL is a product. The processing required
in the step depends upon the liquid content of the inlet gas and desired product
drying, and fractionating the liquids.
2.2.10 SULFUR RECOVERY
Any plant at which H
2
S removal is required, utilizes a sulfur-recovery process if
venting the H
2
S will exceed environmental limits. The common recovery pro-
Chapter 7.
ing attractive. Chapter 8 presents details on nitrogen removal.
brief description of this process is given in Chapter 9. The chapter also discusses
Chapter 7 discusses turboexpanders coupled with compressors, and Chapter 4
slate. Chapter 10 covers the common processing components for sweetening,
cesses, including tail gas cleanup, are discussed in Chapter 11.
© 2006 by Taylor and Francis Group, LLC
Overview of Gas Plant Processing 29
2.2.11 STORAGE AND TRANSPORTATION
of natural gas and NGL products.
2.2.12 LIQUEFACTION
As noted earlier, liquefaction is not common in the United States, but is becoming
more important worldwide as more gas is imported to industrial nations from
remote locations. Also, liquefaction plays an increasingly important role as a
the unique properties of liquefied natural gas (LNG), the chapter includes storage,
transportation, and vaporization processes as well. These aspects are important
to countries, such as the United States, that import LNG.
2.3 IMPORTANT SUPPORT COMPONENTS
Some important components of all gas plants are omitted from detailed discussion
in this book. These components include utilities, process control, and safety sys-
tems. The Engineering Data Book (2004a) provides more details on these topics.
2.3.1 UTILITIES
Utilities include power, heating fluids (steam and oil), cooling water, instrument
air, nitrogen-purge gas, and fuel gas. Most gas plants purchase electrical power, but
some generate at least part of their power on site. Cogeneration plants are becoming
more attractive options for reduction of operating costs, especially when gas tur-
bines are used for driving compressors. Compared with refineries and most chemical
complexes, steam and hot oil are not extensively used in gas plants. Their primary
uses are for regeneration of solvents and some reboilers. Cooling water is used
primarily in heat exchangers on compressors. The Engineering Data Book (2004c)
has an excellent discussion on water treating chemistry.
An uninterrupted source of clean, dry instrument air is critical to plant oper-
ations because the air drives most automated valves. Typical pressures are around
100 psig (7 barg). Plants use one or more backup compressors to ensure that air
to avoid potential line freezes in cold weather. In areas of high instrument flow
rates, air receivers permit large flow rates without pressure drops (Engineering
Data Book, 2004c).
Nitrogen is used as a purge gas around rotating seals, as well as to purge and
blanket vessels. The required purity depends upon the application but usually is
not high. In many cases, the enriched nitrogen is obtained from membrane or
volumes are required, cryogenic fractionation of air is the most economical
process (Engineering Data Book, 2004c).
Although not shown in Figure 2.1, one other “process” tied in with gas plants is
storage and transportation. Chapter 12 discusses the storage and transportation
means for gas storage. Chapter 13 discusses liquefaction processes. Because of
is always available. Many operations have molecular sieve driers (see Chapter 6)
pressure swing adsorption (PSA) separation (see Chapter 9 for details). If large
© 2006 by Taylor and Francis Group, LLC
30 Fundamentals of Natural Gas Processing
Gas plants use the gas they process to fuel the facilities. Boilers, hot-oil fur-
naces, and reciprocating compressors can use low pressure gas. The primary con-
cern is having a particulate-free gas and a roughly constant heating value. The
Engineering Data Book (2004c) states that for gas turbines, which are discussed in
2.3.2 PROCESS CONTROL
Process control has always played a role in gas plants but has become more
important over the years as companies try to reduce labor costs. Most plants use
good digital control systems (DCS) for individual units to provide both process
control and operations history. Since the 1990s, advanced process control (APC)
systems, which “sit” on top of existing DCS systems, provide sophisticated plant
control. APC uses multivariable algorithms that are trained in the plant to optimize
operations. Another aspect of process control commonly used is SCADA (super-
visory control and data acquisition). One important use of SCADA is the monitoring
of field operations, with the capability of controlling dehydration equipment, flow
valves, and compressor stations from the gas plant.
Automation requires accurate input data to make the proper control decisions.
Plants usually have a full-time instrument technician to maintain and calibrate
the many temperature, pressure, and flow sensors, as well as instruments required
for compositional and trace-component analysis.
2.3.3 SAFETY SYSTEMS
Safety systems are critical to all gas plants. These systems include the emergency
shutdown of inlet gas, as well as relief valves and vent systems leading to the
flare. The Engineering Data Book (2004b) provides criteria for sizing relief
systems and flares. Proper sizing of relief valves, rupture disks, and piping is
important to ensure that operating systems can be vented quickly. The design
often is complicated by the need for two phase flows through valves and lines.
Pipe flares are probably the most common flare in gas plants. In normal
operations, the flare flame is barely detectable. If the plant is venting mostly
methane, the flare flame is bright but smokeless. If methane plus heavier hydro-
carbons are flared, the flame will smoke. To make the flare smokeless, steam or
high-flow-rate air or fuel gas is added. If the fuel has a low Btu content (e.g., tail
gas from a sulfur recovery unit), fuel gas is added to ensure complete combustion.
2.4 CONTRACTUAL AGREEMENTS
AND ECONOMICS
When operational changes are under consideration, the customary analysis for opti-
mization of the balance between capital expenditures and operating costs applies.
However, contractual agreements complicate a gas plant economics study whenever
the producer and processor are not the same company. Five basic kinds of contracts
are commonly used between producers and processors (Kuchinski, 2005):
Chapter 4, the required fuel-gas pressure may be as high as 600 psia (41 bar).
© 2006 by Taylor and Francis Group, LLC
Overview of Gas Plant Processing 31
1. Fee based
2. Percentage of proceeds
3. Wellhead purchase
4. Fixed efficiency
5. Keep whole
2.4.1 FEE-BASED CONTRACTS
In fee based contracts, the producer pays the gas processor a set fee on the basis
of gas volumes produced. The processor may obtain additional income by charg-
ing fees for additional services, such as gathering, field compression, pipeline
transmission, and marketing. In these contracts, the processors income is inde-
pendent of gas and NGL prices.
2.4.2 PERCENTAGE OF PROCEEDS CONTRACTS
In percentage of proceeds (POP) contracts, the two parties agree to what percentage
of the proceeds from the sale of the gas and liquids is to be retained by the producer.
Typically, the producer retains more than 70% of the proceeds from the sale of all
products. In the case of multiple producers, each has a percentage share of the
proceeds, allocated on the basis of each producer’s contribution to the proceeds.
Allocations are computed on the basis of the Btu content of the gas delivered at
the wellhead for a producer divided by the sum of the Btu content of the gas from
all producers. Producers and processors share the effect of gas and NGL price
fluctuations.
2.4.3 WELLHEAD PURCHASE CONTRACTS
In wellhead purchase contracts, the processor executes a contract to purchase total
Btus from the producer at a negotiated price usually based against an index. This
purchase is a straight-forward purchase, and the processor’s profits depend upon the
cost of gathering and production and the selling price of the gas and liquids.
2.4.4 FIXED EFFICIENCY CONTRACTS
In fixed efficiency contracts, the processor agrees to provide a certain percentage
recovery (efficiency) of the heavier than methane components from the gas and to
pay the producer on the basis of the market value of the theoretical liquid production
and resultant residue gas. The processor makes money by processing at a higher
efficiency (higher fraction of the liquids removed from the feed stream). Processor
profits hinge on actually having higher recoveries and a favorable price margin.
2.4.5 KEEP WHOLE CONTRACTS
In keep whole contracts, the processor agrees to process or condition the pro-
ducer’s gas for sale in the natural gas market and to return to the producer 100%
of the Btu content of the raw gas (keep the producer whole on Btus) in exchange
© 2006 by Taylor and Francis Group, LLC
32 Fundamentals of Natural Gas Processing
for retaining ownership of all liquids extracted from the gas. The processor usually
retains all liability for fuel, processing costs, and the purchase of replacement of
Btus extracted as a liquid product.
These contracts are more complex, more favorable to the producer, and more
risky to the processor. The producer, in essence, sells the whole hydrocarbon
stream, at the price of natural gas on a Btu basis, to the processor. The processor
makes or loses money, depending on the price difference (price margin) between
natural gas and the NGL, which the processor sells.
Most contracts contain penalties for variations from contracted liquid content,
impurities, and delivery pressure. Contracts may be set to allow for incremental
variations from base composition. Contracts are commonly a combination of two
or more of the five basic types.
How costs are shared between producer and processor for capital projects
depends upon the nature of the project and the contract. New capital items that
benefit both parties may be cost shared. However, maintenance, replacements,
and costs of environmentally driven projects are borne by the processor as a cost
of staying in business. Situations arise in which costs are too high for the processor
to absorb alone, and producers must decide to either share the costs or cease
production.
The combination of the varied and complex contractual agreements, the
proprietary nature of economic data, and the sometimes biased data in the liter-
capital cost data, but otherwise, only qualitative economic comparisons are pro-
vided in the other chapters.
briefly describes three gas plants. The plants differ in both feed and product slate.
REFERENCES
Energy Information Administration, Office of Oil and Gas, U.S. Department of Energy,
U.S. LNG Markets and Uses, Department of Energy, January 2003,
Retrieved June 2005.
Engineering Data Book, 12th ed., Gas Processors Supply Association, Tulsa, OK, 2004a.
Engineering Data Book, 12th ed., Sec. 5, Relief Systems, Gas Processors Supply Association,
Tulsa, OK, 2004b.
Engineering Data Book, 12th ed., Sec. 18, Utilities, Gas Processors Supply Association,
Tulsa, OK, 2004c.
Kuchinski, J., private communication, 2005.
www.eia.doe.gov/pub/oil_gas/natural_gas/feature_articles/2003/lng/lng2003.pdf,
ature limits the discussion of economics in this book. Chapter 14 provides some
To show how the various processing components tie together, Chapter 15
© 2006 by Taylor and Francis Group, LLC
33
3
Field Operations and
Inlet Receiving
3.1 INTRODUCTION
Gas plant operators usually take responsibility for the gas after it leaves the
gas−liquid separator at the wellhead because well operations normally are not
under their direct supervision. This chapter touches on wellhead operations and
Field Liquids Removal
Gas & Liquids from Wells
Inlet
Compression
Sulfur
Recovery
Gas Treating
Dehydration
Nitrogen
Rejection
Helium
Recovery
Outlet
Compression
Elemental
Sulfur
Field Dehydration
Field Acid Gas Removal
Field Compression
Sales Gas
Water &
Solids
Natural
Gas Liquids
Natural
Gasoline
Raw Helium
CO
2
Liquefaction
Liquefied
Natural Gas
Hydrocarbon
Recovery
Liquids Processing
Inlet Receiving
© 2006 by Taylor and Francis Group, LLC
34 Fundamentals of Natural Gas Processing
then discusses the field operations and equipment for moving the gas into the gas
plant and then through inlet receiving, the first of many gas−liquid separations
in the typical gas plant. The chapter includes a discussion of gas hydrates because
their control and prevention is critical to field operations.
3.2 FIELD OPERATIONS
3.2.1 WELLHEAD OPERATIONS
Figure 3.1 shows a typical modern gas well, with the wellhead, gas−liquid
flow meter, along with the solar collector that powers electronic transmission of
the data. The wellhead is in the background. Most companies use electronic
metering. However, some wellheads still have circular chart recorders to measure
flow rates. The meter is where the gas processor purchases the gas from the
producer and the lease royalty owner. Often, the producer and gas processor
belong to the same company. If the gas is nonassociated, hydrocarbon liquids
knocked out in the separator may be remixed with the gas or stored in a tank and
removed by truck. Usually, no compression is applied before the gas is sold.
If the gas is being stripped from oil, separators knock out both oil and water.
For high-pressure wells, the oil passes through up to three separators to recover
the light ends. Because the last-stage separator is near ambient conditions, the
gas may be compressed before it flows into a gathering line to the gas plant.
Frequently, several wells from one lease are tied to one separations unit to
reduce the number of separators, compressors, and meters. This practice is especially
FIGURE 3.1 Gas well with condensate tank and separator.
separator, instrument shed, and condensate tank. Figure 3.2 shows the electronic
© 2006 by Taylor and Francis Group, LLC
Field Operations and Inlet Receiving 35
true in offshore operations because producers need to minimize the number of
platforms and platform weight for cost purposes. Gas dehydration and sometimes
“dew pointing,” that is, reduction of the amount of heavier hydrocarbons, is done
respectively.
3.2.2 PIPING
For offshore production, many wells are tied back to a platform, and then gas
from multiple platforms are tied together into large pipelines that go to the gas
plant, which usually is onshore. Pipelines from onshore wells, especially if
operated by small independent producers, form an extensive network of small
lines from individual wells that tie into increasingly larger lines. The smaller
gathering lines may be aboveground or buried, whereas larger lines are always
FIGURE 3.2 Electronic flow-measurement transmitting equipment. (Courtesy of The
Williams Companies, Inc. of Tulsa, Oklahoma.)
at the lease. Chapters 6 and 7 discuss dehydration and dew-pointing methods,
© 2006 by Taylor and Francis Group, LLC
36 Fundamentals of Natural Gas Processing
buried. Aboveground lines are much easier to maintain but are exposed to the
atmosphere. Surface coating for corrosion prevention and possibly insulation, may
be required.
For low-pressure gathering systems with small amounts of liquid present, the
processor commonly uses a “drip system.” This system involves burying a vessel,
frequently a larger-diameter pipe, below low points in the gathering line. Liquids
drain from the gathering line into the vessel. The vessel is emptied periodically
and the liquid is trucked to the gas plant for processing (McCartney, 2005).
As onshore gas fields are being depleted, wellhead and line pressures are
becoming increasingly subambient. Subambient lines are a major concern for gas
is an unwelcome component because it enhances corrosion, adversely affects several
plant processes, and can cause the sales gas to be subquality if the concentration
exceeds 1 vol% (Engineering Data Book, 2004a). The oxygen problem is severe
enough in some areas that personnel are dedicated to “chasing air” to find the leaks.
They analyze pipeline contents for oxygen and work back into the field to find the
source. Once found, the line or well is shut down for repairs. Since the late 1990s,
gas processors have installed SCADA (supervisory control and data acquisition)
systems to monitor both plant and field operations. These systems provide gas plant
operators with the capability to remotely shut down wells and close lines, which
may be in desolate country, miles from the plant.
3.2.3 COMPRESSION STATIONS
Figure 3.3 shows an onshore compression station, commonly called a booster
(Booster stations are rarely used in offshore production unless the line is roughly
FIGURE 3.3 Booster station on gathering system. (Courtesy of The Williams Companies,
Inc. of Tulsa, Oklahoma.)
station. Figure 3.4 is a schematic of common operations at a booster station.
processors because of potential air intake from leaks. As noted in Chapter 1, oxygen
© 2006 by Taylor and Francis Group, LLC
Field Operations and Inlet Receiving 37
100 miles or more offshore.) In addition to compressing the gas to move it into
the gas plant, the stations usually have inlet suction separators (scrubbers) to
remove condensed liquids. The condensates are either reinjected downstream of
the compressor discharge or stored before being trucked to the gas plant. Some
and mitigate hydrate formation. When required, chemical inhibitors for corrosion
and gas hydrate prevention are injected into the discharged gas; metering pumps
are used to control injection rates.
How to handle condensed liquids at booster stations is an economic decision.
Liquids collection requires tankage and trucking to move the liquids to the plant.
The benefits are higher line capacity, reduced corrosion rates, and reduced chem-
ical hydrate and corrosion inhibitor injection rates. Also, pigging requirements
(see next section) are reduced, as well as the potential for large liquid slugs
entering the plant. Putting liquids back into the pipeline eliminates the need for
tanks and trucking but increases chemical costs, puts more importance on good
pigging, and probably increases pipeline maintenance costs.
Calculation of the pressure drop in these two-phase lines is complex and not
covered here. The Engineering Data Book (2004c) discusses two phase flow and
gives some calculation methods. However, the best results are obtained from
computer models.
Compressor power requirements depend upon gas flow and pressure ratio
Compressors are driven by electric motors, internal-combustion engines, and gas
turbines. Horsepower requirements and availability of fuel dictate the best choice.
Some compressor drivers are internal-combustion engines that have been in
service for 50 or more years and have been rebuilt many times. Compressor
engines are usually fueled with natural gas. The best fuel is sales gas from the
FIGURE 3.4 Schematic of common operations at a compression station on a gathering
system. Condensed liquids may be stored or returned to line. Some locations may have
dehydration or sweetening facilities.
Inlet scrubber Compressor Air cooler
Metering
pump
Corrosion
inhibitor
Hydrate
inhibitor
Water/condensate
storage
Gas to plant
Gas from
wells
booster stations include dehydration facilities (see Chapter 6) to reduce corrosion
(see Chapter 4). Booster-station horsepower ranges from the teens to the thousands.
© 2006 by Taylor and Francis Group, LLC
38 Fundamentals of Natural Gas Processing
plant because it is clean and has a constant heating value. However, this type of
fueling requires running pipe between the plant and booster stations. Raw gas
can be used, but may be too rich for the engines. Fuel-gas conditioners employ
low-temperature separators (LTS) to condense heavier hydrocarbons from the gas
combustion engines, the timing of the engine must be retarded to prevent deto-
nation problems (Kuchinski, 2005). Electric motors are often used in remote sites
when the pipeline gas cannot be used or where emissions from a gas engine
burning rich fuel are an environmental concern.
3.2.4 PIGGING
Pigging is the process of forcing a solid object through a pipeline. Two papers
by Webb (1978a, 1978b) provide an extensive discussion of pigging for both
liquid and gas pipeline. The process involves inserting the pig, via a pig launcher,
into the pipelines and removing it by use of a pig receiver. Pigging is used to
perform any of the following functions:
1. Provide a barrier between liquid products that use the same pipeline
2. Check wall thickness and find damaged sections of lines
3. Remove debris such as dirt and wax from lines
4. Provide a known volume for calibrating flow meters
5. Coat inner pipe walls with inhibitors
6. Remove condensed hydrocarbon liquids and water in two phase pipelines
For field operations, the last function is the most important. Gathering systems
typically are in the two phase stratified flow region, where the liquid flow
rate is much slower than the gas flow rate. Thus, liquid accumulates in low
spots in the line. Field operations must follow a rigorous pigging schedule
to prevent the plant from being hit by large slugs of liquid that would flood
inlet receiving and carry liquids into the gas-processing units. Fortunately,
plant operators usually know when a “killer pig” is coming, and they draw
down liquid levels in inlet receiving. To protect the plant from large liquid
surges, operators respond by shutting in gas, which shuts down field com-
pressors and upsets the plant with the potential for producing off-spec products.
Varnell and Godby (2003) discuss procedures to minimize the potential for
these problems.
Pigs are usually made of polyurethane foam. The smaller, projectile-shaped pigs
in the front of the picture and the larger ones with similar shape are used for
cleaning out soft deposits, gauging lines, and removing water after a line is
hydrotested. The pig in the center of the back row has metal brushes for removing
hard solids. The back two pigs on the left are designed for bidirectional flow.
Larger pigs have a metal mandrel holding the cups and disks, which can be
replaced after wear (Girard, undated).
Figure 3.5 shows examples of pigs and spheres used in two phase service.
and lower the heating value (see Chapter 7). If rich gas is used with internal
© 2006 by Taylor and Francis Group, LLC
Field Operations and Inlet Receiving 39
Foam spheres, like those shown in the left-front of Figure 3.5, are more
commonly used in gathering lines because multiple spheres can be loaded into a
pig launcher for remote launching. Spheres for nominal 2 to 4 inch (50 to 100 mm)
diameter lines are solid, whereas larger ones are inflated, usually with water or
ethylene glycol–water mixtures.
Spheres are launched from a pipe branch that can be isolated and purged
for loading. Figure 3.6 shows a sphere launcher. The launchers typically hold
three to five spheres and are inclined so that spheres can be inserted remotely by
FIGURE 3.5 Different shapes of foam spheres and pigs used to clean gathering lines.
(Courtesy of Girard Industries.)
FIGURE 3.6 Remote automated sphere launcher. (Courtesy of Pearl Development
Company.)
© 2006 by Taylor and Francis Group, LLC
40 Fundamentals of Natural Gas Processing
gravity feed. At the end of the run, the sphere is caught in a sphere catcher
(usually called a pig catcher even if spheres are used). Both launchers and
receivers are on the straight run of a pipe tee. Gas released during venting,
before emptying of the catcher, goes to low-pressure fuel lines or is recom-
pressed for processing (Webb, 1978b).
Pressure differential drives the sphere through the line. However, it usually
stops intermittently when it comes to low points where pools of liquid accumulate.
Field and plant operators learn the typical run times for spheres in a line, and in-
line detectors signal a sphere’s location.
3.3 GAS HYDRATES
A field of technology, called flow assurance, exists for ensuring that hydrocarbons
flow unimpeded by line blockage from wells to the point of processing. The three
areas of concern are:
• Wax and asphaltene solids deposition
• Scale (inorganic salt) deposition
• Gas hydrate solids formation
For gas gathering systems, wax and asphaltene deposition is normally not
a serious problem and can be remediated by pigging. Scale is a common issue
at the wellhead but should not be one in the pipeline. However, gas hydrates
strike fear into the hearts of flow assurance people for two reasons. First, hydrate
plugs can occur within minutes without prior warning. The other solids take
weeks, months, or years to cause plugging and are usually detected by increased
line-pressure drop. Second, although hydrate formation can be inhibited in a
number of ways, injection pump failure, separator failure, and process upsets can
suddenly make pipeline contents vulnerable to hydrate formation. Sloan (2000)
provides a good overview of hydrate problems and how to address them,
especially for offshore situations, where hydrate plugging can be a major problem.
3.3.1 PROPERTIES
Gas hydrates are a class of solid, nonstoichiometric compounds called clathrates.
They form when a host material, water for hydrates through hydrogen bonding,
forms a caged structure that contains guest molecules, such as methane. Both
host and guest must be present for the solid to form, but not all of the cages will
be occupied. Gas hydrates should not be confused with salt hydrates, which form
stoichio-metric compounds. For details of hydrate structure, composition, and
phase behavior see, for example, Sloan (1998) and Carroll (2003).
Gas hydrates were a laboratory curiosity until Hammerschmidt (1939) noted
temperatures well above the freezing point of water. Figure 3.7 shows a break
in all of the curves at 32°F (0°C) because the hydrates are in equilibrium with
that they caused pipeline plugging. As Figure 3.7 shows, hydrates form at
© 2006 by Taylor and Francis Group, LLC
Field Operations and Inlet Receiving 41
gas and ice at the lower temperatures, and with gas and liquid water at the higher
temperatures. For the ethane hydrate line, an abrupt change occurs at 57°F (14°C).
At this point, the hydrate is in equilibrium with gaseous ethane and both liquid
ethane and liquid water (i.e., the hydrate formation line intersects the vapor
pressure line, which is not shown). At higher temperatures, the ethane hydrate is
in equilibrium with liquid ethane and liquid water. The same explanation applies
to the propane. For gas-processing applications, the intersection with the vapor-
pressure line puts an upper temperature limit on hydrate formation because the
line is essentially vertical at pressures commonly seen in gas processing. No
similar point exists for methane, because methane has a critical temperature of
with the predicted hydrate formation conditions for the Rio Arriba County, New
Mexico, gas mixture (see Table 1.4). This mixture contains only 1.33 mol% ethane
and 0.19 mol% propane, but the hydrate formation pressure at a given temperature
drops by 30% compared with pure methane. At a given pressure, pure methane
forms hydrates at the lowest temperature. Adding ethane through butane, hydrogen
sulfide, or carbon dioxide raises the formation temperature significantly. However,
nitrogen will lower the hydrate formation temperature of the mixture. Heavier
compounds, C
5
+, are too large to fit into the cages formed by natural gas and
have an insignificant effect as a diluent because their concentration is so low.
Helium also acts as a diluent.
FIGURE 3.7 Hydrate formation conditions for pure methane, ethane, and propane (data
from Sloan, 1998) and natural gas mixture with composition equal to that of Rio Arriba
CSMHYD98. See text for details.
10
100
1000
−10 10 0 20 30 40 50 60
Temperature, °F
P
r
e
s
s
u
r
e
,

p
s
i
a
Propane
Natural gas mixture
Ethane
Methane
County, New Mexico. Mixture composition given in Table 1.4; line computed using
Figure 3.7 also illustrates the compositional dependence on hydrate formation
about −180°F (−82°C) (see Appendix B).
© 2006 by Taylor and Francis Group, LLC
42 Fundamentals of Natural Gas Processing
3.3.2 HYDRATE FORMATION PREDICTION
Thermodynamics provides a powerful tool for prediction of the temperature and
pressure for hydrate formation on the basis of gas composition. However, even when
hydrates are thermodynamically possible, they may never form. Hydrate formation
kinetics is complex and poorly understood, partly because the crystal growth process
is random. To be safe, operating conditions should be outside the hydrate region.
Since the 1970s, a method based upon statistical thermodynamics has been used
to predict formation temperatures to within a few degrees Fahrenheit. It is the most
successful and widely used application of statistical thermodynamics in industry. The
model is complex and requires a computer program to use. However, commercial
simulators and stand-alone programs are available that use improved models, incor-
porate the effect of thermodynamic inhibitors, and accurately predict phase behavior
for all phases present. A book by Sloan (1998) includes a program, CSMHYD98,
that uses the statistical thermodynamic model, along with a guide to its use.
Before the statistical mechanical model, two empirical methods were widely used.
One employed K-values similar to those used for vapor−liquid equilibrium calcula-
tions, and the other is based upon gas gravity. The K-value method gives slightly
better results for sweet natural gas with specific gravities between 0.6 and 0.8, whereas
it is much better for sour gases. We discuss only the gas gravity correlation, which
is adequate for quick calculations. (The Engineering Data Book [2004d] provides
details of both empirical methods.) If more accurate values are needed, computer
programs that employ the statistical mechanical model should be used.
function of gas specific gravity. For below 1,000 psi (70 bar), the figure can be
approximated by
t(°F) = −16.5 − 6.83/(SpGr)
2
+ 13.8 ln[P(psia)] (3.1a)
t(°C) = −6.44 – 3.79/(SpGr)
2
+ 7.68 ln[P(bara)] (3.1b)
The specific gravity is defined as the ratio of the mass of a given volume of
a gas to that of an equal volume of air; both volumes are measured at 14.7 psia
and 60°F (1.01 bar and 15.6°C). For an ideal gas, the specific gravity is molar
mass of the gas divided by the molar mass of air (28.96).
Example 3.1 Estimate the hydrate-formation temperature at 325 psia (22.4 bar) for
Equation 3.1.
Use of either Figure 3.8 or Equation 3.1 requires knowledge of the specific gravity
gravity = molar mass
gas
/molar mass
air
= 20.08/28.96 = 0.693.
By use of Equation 3.1, we obtain
t = −16.5 − 6.83/(0.693)
2
+ 13.8 × ln(325) = 49°F. Figure 3.8 gives 50°F, and the
experimentally reported value for a 0.7-gravity gas is 50°F (Engineering Data Book,
the gas with the composition in Table 3.1. Compare the results from Figure 3.8 and
of the gas, so this value is calculated first, as shown in Table 3.1. The specific
Figure 3.8 shows hydrate formation prediction curves for natural gases as a
© 2006 by Taylor and Francis Group, LLC
Field Operations and Inlet Receiving 43
FIGURE 3.8 Pressure−temperature curves for estimation of hydrate-formation conditions
as a function of gas gravity. These curves should be used for approximation purposes only.
(Adapted from Engineering Data Book, 2004d.)
TABLE 3.1
Gas Composition for Example 3.1
Component Volume Fraction
a
Molar Mass
lb/lb-Mol Mixture
(kg/kg Mol Mixture)
CO
2
0.002 44.01 0.09
N
2
0.094 28.01 2.63
C
1
0.784 16.04 12.58
C
2
0.060 30.07 1.80
C
3
0.036 44.10 1.59
iC
4
0.005 58.124 0.29
nC
4
0.019 58.124 1.10
Totals 1.000 20.08
a.
For ideal gases volume, fracion is the same as mole fraction
30 40 50 60 70 80 90
Temperature, °F
6000
4000
3000
1500
1000
800
600
400
300
200
150
100
80
60
40
0
.
6
G
r
a
v
i
t
y

g
a
s
0
.
7
0
.
8
H
y
d
r
a
t
e

f
o
r
m
a
t
i
o
n

p
r
e
s
s
u
r
e
,

p
s
i
a
M
e
t
h
a
n
e
© 2006 by Taylor and Francis Group, LLC
44 Fundamentals of Natural Gas Processing
2005d). The computer program CSMHYD98 predicts hydrates form at 49°F. The agree-
ment is exceptionally good between the methods and experimental data for this example.
Gas hydrates form anywhere in a pipeline or process stream, but they are
particularly likely to form downstream of orifices or valves because of Joule-
Thomson expansion effects. The Engineering Data Book (2004d) provides charts
for estimation of how much pressure drop is allowed to avoid hydrate formation
during expansion. However, the charts are only estimates, and calculations by a
computer program should be used to determine the possibility of hydrate formation.
3.3.3 HYDRATE INHIBITION
Three ways exist to avoid hydrate formation in natural gas streams:
• Operate outside the hydrate formation region.
• Dehydrate the gas.
• Add hydrate inhibitors.
The first option is ideal but usually impractical. If practical and economically
feasible, dehydration is the next best option. Because of the high gas volumes
but it is less common in onshore field operations. The third best, but the most
commonly used option, is addition of inhibitors. Whenever operating conditions
are prone to hydrate formation and liquid water may form in the lines, field
operations control hydrate formation by use of chemical inhibitors. They are
added at the wellhead and at booster stations through positive displacement pumps
so that the addition rate can be accurately controlled.
Regarding dehydration, how dry a gas should be to prevent hydrate formation
is uncertain. The gas obviously should have a water dew point below the lowest
operating temperature to avoid water condensation. Thermodynamics predicts that
hydrates can form even when the gas phase is unsaturated with water. This problem
has occurred in several instances in pipelines in which the gas residence time
(Sloan, 2005) was unusually long, but it is extremely rare. This occurrence is
explainable through mechanistic arguments. Nominally, six water molecules are
required for each guest molecule. However, 20 or more water molecules are needed
to form a cage around a gas molecule, and many cages must combine to form the
hydrate lattice. Therefore, the likelihood of sufficient number of water molecules
at concentrations of parts per thousand coming together to form the lattice is low.
Use of chemical inhibitors is the least attractive hydrate inhibition method
for several reasons:
• The proper inhibitor dosage must be known to avoid plugging or
needless chemical costs, but oftentimes it is determined empirically.
• The chemical cost, although it is usually a small fraction of overall
operating costs.
and high pressures, offshore operations frequently use dehydration (see Chapter 6),
© 2006 by Taylor and Francis Group, LLC
Field Operations and Inlet Receiving 45
• The reliability of inhibitor injection can be a problem because of
malfunctioning injection pumps and depleted inhibitor reservoirs, espe-
cially at remote sites.
• The possible interaction between hydrate inhibitors and other additives
reduces the effectiveness of some of additives, an effect that is usually
determined empirically.
Despite these distractions, multimillions of dollars are spent each year on hydrate
inhibitors because they require minimal capital cost to implement and because
injection rates are readily altered according to flow and operating conditions.
Chemical hydrate inhibitors come in three types:
1. Antiagglomerates (AA)
2. Kinetic (KHI)
3. Thermodynamic
The first two types are technologies developed in the late 1990s. Antiagglom-
erates prevent small hydrate particles from agglomerating into larger sizes to
produce a plug. The inhibitors reside in the liquid hydrocarbon phase and are
most often used in pipelines where gas is dissolved in oil. They require testing
to ensure proper concentrations.
Kinetic inhibitors slow crystal formation by interfering with the construction
of the cages. Their advantage is that they can be used at concentrations in the
1 wt% range in the aqueous phase, and they are nonvolatile. Their disadvantage
is that the proper dosage must be determined empirically, as too much inhibitor
may enhance hydrate formation rates. These inhibitors are limited to a subcooling
(difference between desired operating temperature and hydrate formation tem-
perature at constant pressure) of 28°F (15.5°C). However, Mehta and Klomp
(2005) recommend a maximum subcooling of 20°F (11°C) for these kinds of
inhibitors. Kinetic inhibitors are being used in many offshore operations and will
become more widely applied as experience with their use increases.
Thermodynamic inhibitors, mainly methanol and ethylene glycol, are widely
used. They are essentially antifreeze. Inorganic salts are effective but rarely used,
and further discussion relates only to methanol and ethylene glycol.
The required dosage of thermodynamic inhibitors is predictable, but the
concentrations can be high, over 50 wt% of the water phase. A number of
empirical correlations, on the basis of thermodynamic properties of solutions,
predict the amount of any hydrate inhibitor required to depress hydrate formation
temperatures. The two most commonly used are discussed here (more rigorous
methods are used with the computer models discussed above). Hammerschmidt
(1939) proposed the following equation:
(3.2a) ∆t F
X
MW X
i
i i
( )
( )
° =

2335
1
© 2006 by Taylor and Francis Group, LLC
46 Fundamentals of Natural Gas Processing
or
, (3.2b)
where t is the hydrate-depression temperature,°F, X
i
is the mass fraction of
inhibitor in the free-water phase, and MW
i
is the molecular weight of the inhibitor.
The Engineering Data Book (2004d) recommends use of this equation for methanol
concentrations of 20 to 25 wt% of the water phase; it recommends the equation
for ethylene glycol concentrations up to 60 to 70 wt%. However, it notes that the
constant 2,335 may be too low in field situations, and a constant as high as 4,000
may be appropriate. The smaller figure is best used first, and then lower concen-
tration levels used on the basis of operating experience.
Nielsen and Bucklin (1983) proposed
∆t(°F) = −129.6 ln x
W
, (3.3)
where x
W
is the mole fraction of water in the aqueous phase. The paper
provides details on the equation’s derivation and gives example calculations. It
also discusses methanol injection systems. The Engineering Data Book (2004d)
recommends use of this equation for methanol concentrations up to 50 wt%.
3.3.3.1 Methanol vs. Ethylene Glycol
Methanol is more widely used than ethylene glycol, and both have advantages and
disadvantages, primarily on the basis of their physical properties. Both inhibitors
are hydrophilic and remain predominantly with a condensed water phase, even if
a condensed hydrocarbon phase is present. However, methanol is volatile. Because
the gas volume greatly exceeds the water volume in most gathering systems,
a means for estimation of vaporization losses and losses in the hydrocarbon phase.
For ethylene glycol, vaporization losses are negligible, and a solubility loss of 0.3
lb/1,000 gal (0.07 kg/m
3
) of NGL is a common estimate for design purposes
(Engineering Data Book, 2004d). These inhibitor losses to the condensate assume
the condensate is low in aromatics. If aromatic concentrations are atypically high,
inhibitor losses can be significantly higher.
Ethylene glycol is relatively easy to recover, as it remains with the aqueous
phase and can be concentrated by evaporation of the water. Separation of methanol
from water requires distillation. Some gas plants have recovery facilities, but
most ship the water−methanol mixtures elsewhere for recovery or disposal. For
onshore operations, the methanol content in condensate typically presents little
or no problems. However, refineries impose a penalty fee for high concentrations
of methanol in hydrocarbon liquids that arrive from offshore Gulf of Mexico
because they must remove the methanol before processing the liquids.
X
t F MW
t F MW
i
i
i
=
°
° +


( )
( ) 2335
methanol vaporization losses must be considered. Figures 3.9 and 3.10 provide
© 2006 by Taylor and Francis Group, LLC
Field Operations and Inlet Receiving 47
The high vapor pressure of methanol also presents a safety hazard. This
problem is a concern for offshore operations where large quantities of methanol
must be stored. Both inhibitors are toxic and restrictions apply to their disposal,
both on land and at sea. One quirk of ethylene glycol is that it interferes with
the oil and grease test performed on water to be discharged from offshore plat-
forms, giving false high values.
Ethylene glycol is viscous and must be either diluted, usually with water, or
shows the viscosity as a function of temperature and concentration). Viscosity is
not an issue with methanol.
Example 3.2 A sweet gas with a specific gravity of 0.73 leaves a gas−liquid
separator at 100°F and 600 psia saturated with water. The gas drops to 35°F before
reaching the next booster station at 500 psia. Assume no hydrocarbon condensate
has formed in the line. Calculate how much methanol must be added to prevent
hydrate formation between the separator and the booster station per MMscf.
Repeat the calculation for ethylene glycol, which is added in an 80 wt% mixture
with water.
ature at 600 psia is 59°F. This value means a 59°F – 35°F = 24°F subcooling
into the hydrate region, and hydrate formation is probable without inhibition.
FIGURE 3.9 Ratio of methanol-vapor composition to methanol-liquid composition.
(Adapted from Engineering Data Book, 2004d), gpsa fig 20-65.
2.4 2.2 2.0 1.8 1.6 1.4 1.2 1.0 0.8
1,000
500
100
65°F
60°F
55°F
50°F
45°F
40°F
35°F 30°F
25°F 20°F
10°F
0°F
MeOH in vapor phase, Lb/MMscf
P
r
e
s
s
u
r
e
,

p
s
i
a
5,000
0.6 2.6
MeOH in water phase, Wt%
Use of Equation 3.1a or Figure 3.8 shows that the hydrate-formation temper-
kept in a warm storage vessel in cold weather (Appendix B contains a chart that
© 2006 by Taylor and Francis Group, LLC
48 Fundamentals of Natural Gas Processing
Determination of the inhibitor rate requires:
1. Determination of the amount of liquid water formed
2. Calculation of the required amount of inhibitor in the water phase
3. Calculation of the required amount of inhibitor in the gas phase
1. The water content of the gas leaving the separator is 95 lb/MMscf and 13
95 lb – 13 lb = 82 lb of water per MMscf drops out in the line, assuming
worse-case conditions.
Methanol Requirement
2. To estimate the concentration of methanol, rearrange Equation 3.3 to give
ln x
W
= −t(°F)/129.6 = −24/129.6 = −0.185
x
W
= 0.831 or x
MeOH
= 0.169 mole fraction. This value is (0.169 × 32)/(0.831
× 18 + 0.169 × 32) = 26.6 wt% or 0.362 lb methanol/lb water. Thus, the
methanol needed in the water phase is 0.362 × 82 = 29.7 lb/M Mscf.
3.
and 600 psia, gives 1.22 lb methanol vaporized per wt% methanol in the aqueous
phase, or 1.22 × 26.6 = 32 lb methanol per MMscf. Thus, the vapor phase consumes
more methanol than does the aqueous phase. The total amount of methanol required
is (26 + 32) = 58 lb/MMscf. If a condensate phase was present as well, the
losses estimated by use of Figure 3.10 would have to be added into that phase.
FIGURE 3.10 Solubility of methanol in paraffinic hydrocarbons as a function of temper-
ature at various methanol concentrations. (Adapted from Engineering Data Book, 2004d.)
140 120 80 100 60 40 20 0 −20 −40
0.01
5.0
Temperature, °F
M
e
t
h
a
n
o
l

i
n

h
y
d
r
o
c
a
r
b
o
n

p
h
a
s
e
,

m
o
l
%

0.1
1.0
0.05
0.5
20
Wt%
methanol
70
60
40
35
lb/MMscf at 35°F and 600 psia (see Figure 6.1 to obtain these data). Therefore,
To estimate the methanol in the vapor phase use Figure 3.9, which, at 35°F
© 2006 by Taylor and Francis Group, LLC
Field Operations and Inlet Receiving 49
This amount of methanol will be much less than what goes into the aqueous
and vapor phase in gas lines.
Ethylene Glycol Requirement
2. Use Equation 3.2b with the constant 3,225 and a molar mass of ethylene
glycol of 62 to obtain the mass fraction of pure glycol required:
However, the glycol is diluted to 80 wt%. To obtain the mass of inhibitor
solution added per unit mass of free water present to obtain the desired con-
centration, we use X
0
/(X
i
– X
0
), where X
0
is the weight fraction of inhibitor
in the solution to be added. The amount of inhibitor solution added per pound
of free water initially present is then 0.8/(0.8 – 0.40) = 2.00, and the total
amount of ethylene glycol solution added is 2.00 × 82 = 164 lb/MMscf.
At these conditions, glycol loss into the vapor phase is negligible, so the total
amount of solution required is 164 lb/MMscf.
3.4 INLET RECEIVING
Gas and liquids that enter the gas plant pass emergency shutdown valves, which
isolate the plant from incoming streams and pig receivers, and then go to inlet
receiving, where condensed phases drop out. Gas from inlet receiving goes to inlet
critical because downstream gas processing units rely on a continuous gas stream
free of liquids, even when surges of liquid enter the plant. A slug catcher is a gas−
liquid separator sized to hold the biggest slug a plant will experience. Depending
upon slug catcher design, inlet receiving handles just slugs or combines slug catching
with liquid storage. Kimmitt et al.(2001) detail the process to properly size and
design a slug catcher, and the Engineering Data Book (2004b) provides details on
separator design. This section provides an overview of separator principles and then
discusses two slug catcher configurations.
3.4.1 SEPARATOR PRINCIPLES
Called scrubbers, knockout pots, inlet receivers, and just separators, effective
phase separators protect downstream equipment designed to process a single
phase. It is the critical first step in most processes in gas plants and typically is
a simple vessel with internal components to enhance separation. Bacon (2001)
provides an extensive discussion on designing separators and provides four design
examples. (Bacon (2001) provides a spread sheet which includes the design
X
i
=
×
× +
=
24 62
24 62 2235
0 40 .
compression (if necessary), and the liquids go to storage for processing (see Chapter
10). The initial gas−liquid separation occurs in a slug catcher. Slug catchers are
© 2006 by Taylor and Francis Group, LLC
50 Fundamentals of Natural Gas Processing
calculations on the conference proceedings CD.) In this section, the emphasis is
on gas−liquid separation. Because of its importance in gas processing, a short
section on liquid−liquid separation is included. Finally, typical residence times
are given for both gas−liquid and liquid−liquid separators in various types of
service to check vessel sizing.
3.4.1.1 Gas-Liquid Separation
Separator vessel orientation can be vertical or horizontal. Vertical separators are
most commonly used when the liquid-to-gas ratio is low or gas flow rates are
low. They are preferred offshore because they occupy less platform area. How-
ever, gas flow is upwards and opposes the flow of liquid droplets. Therefore,
vertical separators can be bigger and, thus, more costly than horizontal separa-
tors. Inlet suction scrubbers at compressor stations are usually vertical. Hori-
zontal separators are favored for large liquid volumes or if the liquid-to-gas ratio
is high. Lower gas flow rates and increased residence times offer better liquid
dropout. The larger surface area provides better degassing and more stable
indicates the four areas or types of separation:
Primary separation
Gravity settling
Coalescing
Liquid collecting
Primary Separation—Primary separation is accomplished by utilizing the
difference in momentum between gas and liquid. Larger liquid droplets fail
to make the sharp turn and impinge on the inlet wall. This action coalesces
finer droplets so that they drop out quickly. Although inlet geometries vary,
most separators use this approach to knock out a major portion of the incoming
liquid.
Gravity Settling—Gravity settling requires low gas velocities with mini-
mal turbulence to permit droplet fallout. The Engineering Data Book (2004b)
provides detailed information for settling calculations. This section summa-
rizes useful equations for quick estimation of separator performance. The
calculations assume the droplets to be rigid spheres. The terminal-settling
velocity, V
T
, for a sphere falling through a stagnant fluid is governed by particle
diameter, density differences, gas viscosity, and a drag coefficient that is a
function of both droplet shape and Reynolds number. Here the Reynolds
number is defined as
N
Re
= D
P
V
T
r
g
/m
g
, (3.4)
where D
P
is particle diameter, r
g
is the density, and m
g
is the viscosity. Thus,
calculations for V
T
are an iterative process.
liquid level as well. Figure 3.11 shows a schematic of gas−liquid separators and
© 2006 by Taylor and Francis Group, LLC
Field Operations and Inlet Receiving 51
For large particles (1,000 to ~70,000 micron), the terminal velocity is com-
puted by the equation
, (3.5)
where g is the gravitational constant. This equation, known as Newton’s law,
applies when the Reynolds number is greater than 500.
FIGURE 3.11 Gas−liquid separators. (Adapted from Engineering Data Book, 2004b.)
A
D
B
C
Boot
B
C
A
Gas + liquid
Gas
Hydrocarbon
liquid
Mesh pad
Liquid
Gas
Mesh pad
Vortex
breaker
A = inlet device
B = gas gravity separation
C = mist extraction
D = liquid gravity separation
Horizontal separator
Vertical separator
Water
Gas + liquid
D
V
gD
T
p l g
g
=

1 74 .
( ) ρ ρ
ρ
© 2006 by Taylor and Francis Group, LLC
52 Fundamentals of Natural Gas Processing
If the particle size is too large, excessive turbulence occurs and Equation 3.5
fails. The upper limit is found by use of the equation
(3.6)
with K
CR
= 18.13 and 23.64 for engineering and metric units, respectively,
and is based upon a Reynolds number of 200,000, which is the upper limit for
Newton’s law to hold.
At the other extreme, where the flow is laminar (N
Re
< 2), Stokes’ law applies.
The terminal velocity is
(3.7a)
(3.7b)
where Equation 3.7a is in English units and Equation 3.7b is in SI. Stokes’ law
applies to particles in the 3 to 100 micron range. To find the maximum size particle
in this flow regime, use K
CR
= 0.0080 in Equation 3.6, which corresponds to an N
Re
of 2. Particles smaller than 3 microns will not settle because of Brownian motion.
Unfortunately, droplets that condense from a vapor tend to be in the 0.1 to 10 micron
range; the majority are around 1 micron. Entrained droplets are 100 times larger.
To reduce turbulence, the settling section may contain vanes. They also act
as droplet collectors to reduce the distance droplets must fall.
Coalescing—The coalescing section contains an insert that forces the gas
through a torturous path to bring small mist particles together as they collect on
lists some of the features of the wire mesh and vane pack mist extractors. Cyclonic
devices are proprietary devices, and features are application dependent.
Mesh pads are either wire or knitted mesh, usually about 6 inches (15 cm)
thick, and, preferably, are mounted horizontally with upward gas flow, but they
can be vertical. They loose effectiveness if tilted. Mesh pads tend to be more
effective at mist removal than vane packs but are subject to plugging by solids
and heavy oils.
usually spaced 1 to 1.5 inches (2.4 to 3.8 cm) apart, that force the gas and mist to
follow a zigzag pattern to coalesce the mist into larger particles as they hit the plates.
Coalesced drops collect and flow out the drainage traps in the plates. Although not
as effective at removing small drops, they are ideal for “dirty” service because they
D K
g
P CR
g
g l g
( )
=









µ
ρ ρ ρ
2
1
3
V
gD
T
P g l
g
=
− 1 488
18
2
, ( ) ρ ρ
µ
V
gD
T
P g l
g
=
− 1 000
18
2
, ( ) ρ ρ
µ
the insert. These inserts can be mesh pads, vane packs, or cyclonic devices. Table 3.2
Figure 3.12 shows several elements of a vane pack, which are corrugated plates,
© 2006 by Taylor and Francis Group, LLC
Field Operations and Inlet Receiving 53
will not plug. However, solids can collect on the back edge of the vane and plug the
drainage ports. Liquid then collects and is re-entrained. This problem is resolved by
putting coarse filters upstream of the vane pack (McCartney, 2005).
data are based upon an air−water system and differs from natural gas data because
of density and surface tension. However, Figure 3.13 shows the regions where
TABLE 3.2
Features of Mist Extracting Devices
Wire Mesh Vane Pack
Gas capacity factor, C in
Equation 3.8, ft/sec (m/sec)
0.22 – 0.39 (0.067 – 0.12) Horizontal flow 0.9 – 1.0
(0.27 – 0.30)
Vertical flow 0.4 – 0.5
(0.12 – 0.15)
Droplet efficiency 99 − 99.5% removal of 3- to
10-micron droplets
99% removal of 10- to 40-
micron droplets
Turndown range, %
of design gas rate
30 − 110 Rapid decrease in efficiency
with decreased gas flow
Pressure drop, inches
of water (kPa)
< 1 (0.25) 0.5 to 3.5 (0.12 to 0.87)
Source: Engineering Data Book (2004b).
FIGURE 3.12 Sketch of vane pack mist extractor element with liquid drainage traps.
(Adapted from Engineering Data Book, 2004b.)
Gas flow
Drainage traps
Figure 3.13 shows qualitatively the range for mist pads and vane packs. The
© 2006 by Taylor and Francis Group, LLC
54 Fundamentals of Natural Gas Processing
each demister type is effective. Note that these devices fail to coalesce droplets
below around 0.5 micron, and each has both upper and lower velocity limits. The
lower limit is caused by too low a velocity to force sufficient impinging of the
droplets on the solid surface to provide coalescing. At high velocities, the coa-
lesced droplets are stripped from the solid by the high velocity gas.
Cyclones use centrifugal force to enhance separation of condensed phases
from gas. Their advantage is that they are extremely efficient at high-gas throughput,
which means smaller vessel diameter. However, they tend to have a narrow
operating range, relatively high pressure drop, and difficulties handling liquid
slugs. Sterner (2001) presents details on various cyclone separators as well as a
brief discussion on mist pads and vane packs.
The Engineering Data Book (2004b) and Bacon (2001) provide design
calculations for wire mesh and vane pack coalescing units, but cyclonic device
design is proprietary. Below is a short discussion on the determination of sepa-
ration performance for mesh pads and vane packs.
The two parts of separation performance are droplet-removal efficiency and
gas-handling capacity. Manufacturers provide charts on removal efficiency as a
function of droplet size at design conditions. Gas handling capacity can be
estimated by use of the following equations:
(3.8)
FIGURE 3.13 Approximate operating ranges for different kinds of demisters. Data are
based upon water and air. (Courtesy of ACS Industries, 2005.)
1000
Gravity
settling
Vane
packs
Wire
mesh
pads
R
e
-
e
n
t
r
a
i
n
m
e
n
t
100
10
1
Knit mesh pads
Poor impact efficiency
D
r
o
p
l
e
t

s
i
z
e
,

m
i
c
r
o
n
s
0.1
Gas velocity, ft/sec
1 10 30
V C
t
l g
g
=
− ( )
,
ρ ρ
ρ
© 2006 by Taylor and Francis Group, LLC
Field Operations and Inlet Receiving 55

extractor, A, area is given by
A = Q
A
/V
t
, (3.9)
where Q
A
is the actual volumetric flow rate. The Engineering Data Book
(2004b) tabulates the percentage reduction in C as a function of pressure, but it
can be estimated by
C(ft/sec) = 100 – 1.35 P(psig)
0.42
(3.10a)
C(m/sec) = 100 – 4.12 P(barg)
0.42
, (3.10b)
where P is in psig and barg for Equation 3.10a and Equation 3.10b, respec-
tively. This equation should not be used above 1,150 psig (80 barg).
To remove droplet sizes below 3 microns, filter separators and coalescing
filters are required. They are separate units from the gas−liquid separators dis-
cussed here. These filtration units frequently are used upstream of gas treating
units to prevent hydrocarbon liquids and compressor oil from entering amine
the horizontal unit is more common (Engineering Data Book, 2004b). The units
contain a filter coalescing section, followed by a final mist extraction section.
The first section either knocks out the fine mist or coalesces it into larger
droplets. The filter bundle resembles a heat exchanger bundle, with the gas entering
on the shell side. The gas, with coalesced droplets, leaves the tubes and passes
through a final mist extractor that contains either mist pads or vane packs. Pressure
drop through a clean separator is usually 1 to 2 psi (0.07 to 0.14 bar), and filters
are changed when the pressure drop rises to about 10 psi (0.7 bar).
Coalescing filters are similar to filter separators but normally are vertical and
designed for very high gas-to-liquid ratios. However, they contain only filter
cartridges. Details of these units are proprietary. Manufacturers must be contacted
for details of applicability and operating ranges.
Liquid Collection—The liquid collection section acts as a holder for the liquids
removed from the gas in the above three separation sections. This section also provides
for degassing of the liquid and for water and solids separation from the hydrocarbon
phase. The most common solid is iron sulfide from corrosion, which can interfere
with the liquid−liquid separation. If a large amount of water is present, separators
often have a “boot,” as shown in the horizontal separator, at the bottom of the separator
for the water to collect. The Engineering Data Book (2004b) estimates that retention
times of 3 to 5 minutes are required for hydrocarbon−water separation by settling.
3.4.1.2 Liquid-Liquid Separators
Liquid−liquid separators are similar to gas−liquid separators, and the configura-
tion depends upon whether the separator has two or three phases present. If three
where C is the gas capacity factor given in Table 3.2. The required mist
absorbers and causing foaming problems (see Chapter 5). They are upstream of
Filter separators, like other gas−liquid separators, can be horizontal or vertical;
dehydration units to protect solid adsorbants from amine solution (see Chapter 6).
© 2006 by Taylor and Francis Group, LLC
56 Fundamentals of Natural Gas Processing
screen may be in the liquid section to coalesce the droplets. If a vapor phase is
not present, the mixture enters at one end, goes through a coalescing medium,
and then enters the section where gravity separation occurs.
The same equations described for gas−liquid gravity settling apply. However,
the particles typically follow Stokes law. Liquids present two difficulties usually
not encountered with gas−liquid separations: relatively small differences in den-
sity (use of change in direction provides little or no benefit) and possibly emul-
sions. Both of these factors force longer residence times. The Engineering Data
Book (2004b) provides equations for both horizontal and vertical vessel sizing.
However, for quick sizing, a calculation that bases the volume on suggested
residence times is frequently adequate.
3.4.1.3 Residence Time for Various Separator Applications
This section provides typical residence times for a variety of gas−liquid and
liquid−liquid separations seen in both field and plant operations. The residence
time is simply the volume of the phase present in the vessel divided by the
volumetric flow rate of the phase. Table 3.3 provides typical retention times
liquid separations.
3.4.2 SLUG CATCHER CONFIGURATIONS
This section briefly describes two kinds of slug catchers, manifolded piping and
inlet vessels. Kimmitt et al. (2001) provides detailed design information. The
TABLE 3.3
Typical Retention Times for Gas-Liquid Separations
Type of Separation Retention Time (Minutes)
Natural gas condensate separation 2 – 4
Fractionator feed tank 10 − 15
Reflux accumulator 5 – 10
Fractionation column sump 2
a
Amine flash tank 5 – 10
Refrigeration surge tank 5
Refrigeration economizer 3
Heat medium oil surge tank 5 – 10
b
a
If the fractionator column sump is feeding a downstream fractionator column, it should be sized
as a feed tank (McCartney, 2005).
b
This vessel must have adequate space to allow for expansion of the heat medium from ambient
to operating temperature (McCartney, 2005).
Source: Engineering Data Book (2004b).
for common gas−liquid separations. Table 3.4 gives residence times for liquid−
phases are present, the configuration shown in Figure 3.11 is typical. A coalescing
© 2006 by Taylor and Francis Group, LLC
Field Operations and Inlet Receiving 57
most difficult part of a slug catcher design is the proper sizing. Sizing requires
knowledge of the largest expected liquid slug, as liquid pump discharge
capacity on the slug catcher will be trivial compared with the sudden liquid
influx.
3.4.2.1 Manifolded Piping
One reason piping is used instead of separators is to minimize vessel wall
thickness. This feature makes piping attractive at pressures above 500 psi (35
bar). The simplest slug-catcher design is a single-pipe design that is an
increased diameter on the inlet piping. However, this design requires special
upon the required volume and operating pressure. Also, some designs include
a loop line, where some of the incoming gas bypasses the slug catcher. Primary
separation occurs when the gas makes the turn at the inlet and goes down the
pipes. Liquid distribution between pipes can be a problem, and additional lines
between the tubes are often used to balance the liquid levels. In harp designs,
the pipes are sloped so that the liquid drains toward the outlet.
Gravity settling occurs as the gas flows to the vapor outlet on the top while
the liquid flows out the bottom outlet. Pipe diameters are usually relatively small
(usually less than 48 inches [120 cm]), so settling distances are short.
Because manifolded piping is strictly for catching liquid slugs, demisters are
usually installed downstream in scrubbers. Likewise, liquid goes to other vessels,
where degassing and hydrocarbon−water separation occurs.
TABLE 3.4
Typical Retention Times for Liquid-Liquid Separators
Type of Separation Retention Time (Minutes)
Hydrocarbon−water
Above 35° API hydrocarbon
Below 35° API hydrocarbon
> 100°F (38°C)
80°F (27°C)
60°F (16°C)
3 – 5
5 − 10
10 − 20
20 − 30
Ethylene glycol-hydrocarbon separators (cold separators) 20 − 60
Amine-hydrocarbon 20 − 30
Caustic-propane 30 – 45
Caustic-heavy gasoline 30 − 90
Source: Engineering Data Book (2004b).
pigs to accommodate the change in line size. Figure 3.14 shows a schematic
from the end of a multipipe slug catcher. The number of pipes varies, depending
of typical multipipe harp design for a slug catcher. Figure 3.15 shows a view
© 2006 by Taylor and Francis Group, LLC
58 Fundamentals of Natural Gas Processing
FIGURE 3.14 Schematic of multiple-pipe slug catcher. (Courtesy of Pearl Development
Company.)
FIGURE 3.15 Multipipe slug catcher with liquid volume of 500 Bbls (80 m
3
). Nominal
inlet gas rate is 206 MMscfd (5.8 MMSm
3
/d). (Courtesy of Duke Energy Field Services.)
© 2006 by Taylor and Francis Group, LLC
Field Operations and Inlet Receiving 59
Kimmitt et al. (2001) note several advantages to the pipe design, including
the fact that design specifications are based upon pipe codes instead of vessel
codes. Also, the slug catcher can be underground, which reduces maintenance
costs and insulation costs if the slug catcher would otherwise need to be
heated.
3.4.2.2 Inlet Vessels
These slug catchers, commonly called inlet receivers, are simply gas−liquid
separators that combine slug catching with liquid storage. They are usually
employed where operating pressures are relatively low or where space is a
problem. Horizontal vessels are preferred, unless area is limited (as on offshore
platforms), because they provide the highest liquid surface area. Usually two or
three vessels are manifolded together to permit larger volumes and to allow
servicing of one vessel without plant disruption. Length-to-diameter ratios are
typically 3:1 to 5:1 to maintain a low gas velocity through the gravity-settling
section.
With their large volume, inlet vessels usually act as primary storage for inlet
liquids as well. The vessels can accommodate demisters, so additional separation
may be unnecessary before the gas goes to compression.
3.4.2.3 Comparison of Slug Catcher Configurations
Below is a comparison of piping, vertical vessels, and horizontal vessels for slug
catching according to various considerations:
• Land or surface requirements. If no land constraints apply, piping is
attractive. If constraints are severe, as on offshore platforms, vertical
vessels are preferred. Otherwise horizontal vessels are the best choice.
• Operating pressure. If inlet pressures are greater than about 500 psi
(35 bar), significant savings in material costs can be achieved by use
of the smaller diameter piping slug catcher.
• Gas−liquid separation capability. Horizontal vessels provide the best
separation, whereas piping provides the least because its main function
is to catch liquid slugs. The large liquid surface area of horizontal
vessels provides the best degassing. Piping has the shortest gas resi-
dence time when liquid levels are properly maintained in the vessels.
However, with piping, small diameter gas scrubbers can be used for
demisting.
• Liquid storage. Horizontal vessels can act as primary liquid storage,
whereas liquids from vertical vessels and piping must be sent to another
vessel. Regardless of slug catcher used, liquids will go to low-pressure
flash drums to recover light ends.
© 2006 by Taylor and Francis Group, LLC
60 Fundamentals of Natural Gas Processing
Gas plants routinely receive gas from more than one inlet line. In many cases,
the inlet line pressures differ. To minimize required compression on the gas that
leaves inlet receiving, slug catchers operate at each of the inlet line pressures.
3.5 SAFETY AND ENVIRONMENTAL
CONSIDERATIONS
Some common safety issues involved with field operations include the following:
• Leaking pipelines. Leaks present a danger of fire if an ignition source
is available and of poisoning if the gas contains high hydrogen sulfide
concentrations. Respirators may be required for work in a sour gas
field.
• Plugged pipelines. Plugs can be caused by solids blockage, such as
hydrates or, occasionally, a stuck pig. Clearing a hydrate plug should
be done by depressurization of both sides of the plug to prevent it from
dislodging and becoming a projectile that potentially damages the line.
Dislodging a hydrate plug in subsea lines can be complex. See Sloan
(2000) for more details.
Spheres rarely stick in well-maintained pipelines. The two most common
reasons for sticking are gas bypass and line obstruction. Technical help from
vendors should be sought. One should contact a sphere manufacturer if common
company practice fails.
Other than the liquid slugging problem already mentioned, few safety con-
siderations apply to inlet receiving. One hazardous operation in this area is
retrieval of pigs from the pig receiver.
Another consideration is the impact of liquid slugs on the piping. Often, inlet
slug catching equipment is designed to field codes, which allow thinner piping
than would be used in a gas processing plant. If liquid flows exceed design rates,
the required pipe anchors and wall thickness should be checked to ensure the
unit can handle the increased pipe stresses induced by potentially larger slugs of
liquid (McCartney, 2005).
From an environmental perspective, the most obvious problem is leaking
pipelines. However, if hydrate inhibitors are used, berms are required around the
storage tanks to prevent soil contamination; in the case of methanol, the potential
for fire exists as well. A major issue is control of exhausts from compressors.
Carbon monoxide and NO
X
are the major concerns for the internal-combustion
engines that drive reciprocating compressors, and NO
X
is the main pollutant from
sulfur removal or sulfur oxide emissions may be required.
In situations in which the plant is required to shut-in, gas may have to be
vented or flared at remote booster stations. Rich gas in gathering systems can to
be heavier than air at lower temperatures, which makes venting hazardous. If the
gas is sour, configurations that permit the gas to go to flares may be required.
turbine-driven centrifugal compressors (see Chapter 4). If raw gas is burned,
© 2006 by Taylor and Francis Group, LLC
Field Operations and Inlet Receiving 61
REFERENCES
Bacon, T.R, Fundamentals of Separation of Gases, Liquids, and Solids, Proceedings of
the Laurance Reid Gas Conditioning Conference, Norman, OK, 2001.
Carroll, J.J., Natural Gas Hydrates: A Guide for Engineers, Gulf Professional Publishing,
Houston, TX, 2003.
Engineering Data Book, 12th ed., Sec. 2, Product Specifications, Gas Processors Supply
Association, Tulsa, OK, 2004a.
Engineering Data Book, 12th ed., Sec. 7, Separation Equipment, Gas Processors Supply
Association, Tulsa, OK, 2004b.
Engineering Data Book, 12th ed., Sec. 17, Fluid Flow and Piping, Gas Processors Supply
Association, Tulsa, OK, 2004c.
Engineering Data Book, 12th ed., Sec. 20, Dehydration, Gas Processors Supply Associ-
ation, Tulsa OK, 2004d.
Hammerschmidt, E.G., Preventing and removing gas hydrates formations in natural gas
pipelines, Oil Gas J. 37 (52) 66, 1939.
Kimmitt, R.P., Root, C.R., and Rhinesmith, R.B., Proven Methods for Design and Oper-
ation of Gas Plant Liquid Slug Catching Equipment, Proceedings of the Eightieth
Annual Convention of the Gas Processors Association, Tulsa, OK, 2001.
Kuchinski, J., private communication, 2005.
McCartney, D., private communication, 2005.
Mehta, A.P. and Klomp, U.C., An Industry Perspective on the State of the Art of Hydrates
Management, Proceedings of the Fifth International Conference on Gas Hydrates,
Trondheim, 2005, 1089.
Nielsen, R.B and Bucklin, R.W, Why not use methanol for hydrate control? Hydroc. Proc.,
62 (4) 71, 1983.
Sloan, E.D., Clathrate Hydrates of Natural Gases, 2nd ed. Marcel Dekker, New York, 1998.
Sloan, E.D., Hydrate Engineering, Monograph, Vol. 21, Society of Petroleum Engineers,
Richardson, TX, 2000.
Sloan, E.D., private communication, 2005.
Sterner, A.J., Developments in Gas-Liquids Separation Technology, Proceedings of the
Eightieth Annual Convention of the Gas Processors Association, Tulsa, OK, 2001.
Varnell, B. and Godby, S., Managing Plant Inlet Liquid Slugs, Proceedings of the Eighty-
Second Annual Convention of the Gas Processors Association, Tulsa, OK, 2003, 238.
Webb, B.C., Guidelines set out for pipeline pigging, Oil Gas J., 76 (46) 196, 1978a.
Webb, B.C., More guidelines given for gas liquid pipeline pigging, Oil Gas J., 76 (48) 74, 1978b.
WEB SITES
relate to gas production.
at drilling sites. Not directly related to gas processing.
gives details of current research activities and provides the free downloading of the
hydrate prediction program CSMHYD98 to compute hydrate-formation conditions.
Girard Industries, undated. http://www.girardind.com/index.htm, Retrieved June 2005.
Society of Petroleum Engineers (SPE): www. spe. org/ spe/ jsp/ basic/ 0,,1104 _ 1714 _
1003866,00.html. Site provides history and overview of well-site operations that
Occupational Safety and Health Administration (OSHA): www. osha. gov/ SLTC/ etools/
Colorado School of Mines Hydrate Research Program: www.mines.edu/research/chs/. Site
oilandgas/illustrated_glossary.html: Site provides excellent glossary of terms used
© 2006 by Taylor and Francis Group, LLC
63
4
Compression
4.1 INTRODUCTION
Pressure plays a major role in gas processing, as it moves gas from the field,
through the gas plant, and into the sales gas line. Pressure provides the source
for cooling the gas to low temperatures. Process streams that undergo compression
include the following:
Field Liquids Removal
Gas & Liquids from Wells
Inlet
Compression
Sulfur
Recovery
Gas Treating
Dehydration
Nitrogen
Rejection
Helium
Recovery
Outlet
Compression
Elemental
Sulfur
Field Dehydration
Field Acid Gas Removal
Field Compression
Sales Gas
Water &
Solids
Natural
Gas Liquids
Natural
Gasoline
Raw Helium
CO
2
Liquefaction
Liquefied
Natural Gas
Inlet Receiving
Hydrocarbon
Recovery
Liquids Processing
© 2006 by Taylor and Francis Group, LLC
64 Fundamentals of Natural Gas Processing

the gas must be at 850 to 1,000 psi (60 to 70 bar) when it enters the
hydrocarbon recovery section but at 600 to 650 psi if only primarily
• Gas coming from stabilizing the liquids in liquids processing (see
• Gas from other vapor recovery facilities that collect low-pressure gas.
Low-pressure gas not used as fuel gas is compressed for hydrocarbons
recovery.
• Gas exiting hydrocarbon recovery and entering the pipeline (see
Chapter 7). For gas plants that process more than 5 MMscfd (140
MSm
3
/d) (Crum, 1981), use of turboexpanders to cool the gas for liquid
hydrocarbons recovery is economically attractive. Mechanical energy
generated by the turboexpander provides part of the power for outlet
compression. However, additional outlet compression is required to
bring the sales gas to pipeline pressure.
Typically, the only other demands for compression are in recompressing
adsorption bed regeneration gas and in propane refrigeration in hydrocarbon
recovery. Some plants avoid propane refrigeration by having high inlet gas
pressures. This option is attractive when the inlet gas contains relatively small
amounts of C
2
+ hydrocarbons or when high ethane recoveries are not required.
Gas and refrigeration compressors typically are the largest capital expense
in the construction of a new gas processing plant. They account for up to 50 to
60% of the total installed cost of the facility (McCartney, 2005). They also tend
to have the largest maintenance expense in the facility.
As in the case of pumps, the two broad categories of compressors are positive
displacement and dynamic. Fundamentally, the major difference between the two
categories is the former is a volume displacement device, whereas the latter is a
pressure (or pump head) device, with performance dependent upon flow and outlet
pressure demands. Positive displacement compressors increase pressure by
decreasing volume, whereas dynamic compressors turn gas velocity (kinetic
energy) into pressure. Whereas positive displacement compressor performance is
insensitive to the gas being compressed, dynamic compressors are limited to gases
with molecular weights of about 10 or higher (Jandjel, 2000). This limitation is
not a concern in gas plant service, unless helium-rich gas from a helium-recovery
process needs compression.
After reviewing some fundamentals regarding compression, this chapter pre-
sents an overview of the various types of compressors typically found in gas plants
and lists their strengths and weaknesses. More discussion on compressors other
than reciprocating compressors is presented because they are less familiar to most
engineers. This discussion is followed by ways to estimate power requirements
for a given set of flow conditions or to estimate flow conditions for a given
compressor design. Jandjel (2000) provides a good overview of compressors for
propane and heavier components are recovered (see Chapter 7).
Gas leaving inlet receiving (Chapter 3). To maximize liquids recovery,
Chapter 10).
© 2006 by Taylor and Francis Group, LLC
Compression 65
gas and air service, and the Engineering Data Book (2004a) gives details on power
requirements. Unfortunately, Jandjel (2000) provides no additional references.
4.2 FUNDAMENTALS
This section reviews some basic thermodynamic relations in compression and
then defines some of the common efficiencies used to rate compressors.
4.2.1 THERMODYNAMICS OF COMPRESSION
To compute the reversible shaft work, w
S
, of a continuous open process that
increases the absolute pressure from P
1
to P
2
requires evaluation of the integral
(see a standard thermodynamics textbook [e.g., Smith et al., 2001]).
(4.1)
where all variables are based upon energy per unit mass. Because the gas is
compressible, the integral depends upon the PV path. Minimizing the work
requires minimizing the volume during compression, which implies that the
process must be isothermal. For an ideal gas, where V RT/P, the reversible
isothermal shaft work is given by
w
S
−RT ln(P
2
/P
1
)/MW, (4.2)
where R is the gas constant, MW is the molecular mass, and T is the absolute
temperature. Reversible isothermal work is the minimum work requirement for
a compression process. Note the negative sign indicates that work is being done
on the system and that all pressures are in absolute units.
Achievement of isothermal compression requires infinitely good heat transfer
to remove all heat of compression. This case represents an unrealistic case, and
isothermal work calculations are rarely used in practice.
The other extreme is no heat transfer (adiabatic). All heat generated in com-
pression goes into the gas phase, which maximizes the volume and, thus, requires
the maximum work for compression. Compressors, although operating irrevers-
ibly, more closely follow this path because compression takes place rapidly, with
little time allowed for heat transfer. For reversible adiabatic (isentropic) compres-
sion of an ideal gas, the PV path follows
Pv
g
constant (4.3)
w h vdP
S
P
P
− −


1
2
,
© 2006 by Taylor and Francis Group, LLC
66 Fundamentals of Natural Gas Processing
and the work is calculated by
, (4.4)
where T
1
is the inlet temperature and is the ratio of the molar heat capacities:
g (C
P
/C
V
) [C
P
/(C
P
– R)] (4.5)
The value of g is both temperature and composition dependent. Molar gas
mixture heat capacities are computed by
(4.6)
and C
VM
for natural gas components as a function of temperature. Common practice is
to determine the heat capacity ratio at the average of suction and discharge
temperatures.
Equations 4.2 and 4.4 show that the work required is a function of the ratio of
absolute pressures, sometimes referred to as the “compression ratio.” The actual
pressures have a significant effect on compressor design and cost. They have only
a secondary effect on the work in that they change the heat capacities and g. This
effect is commonly ignored because the calculations provide only estimates. Equa-
tions of state are needed to obtain more accurate work and power requirements.
Example 4.1 Assume an ideal gas, and compute the ratio of heat capacities for the
n-butane and C
5
+ fraction to be hexane (Table 4.1).
TABLE 4.1
Data for Example 4.1
Mol % C
P
at 100°F Btu/lb-mol-°R yC
P
Helium 0.45 4.97 0.022
Nitrogen 14.65 6.96 1.020
Methane 72.89 8.65 6.305
Ethane 6.27 12.92 0.810
Propane 3.74 18.20 0.680
Butanes 1.38 24.32 0.336
Pentanes and heavier 0.62 40.81 0.253
C
PM
9.426
w
RT
MW
P
P
S



|
.

`
}

]
]
]
]

γ
γ
γ γ
1 2
1
1
1
1
( )
( )/
C x C
PM i P
i


is computed the same way. Appendix B gives heat capacity values
southwest Kansas gas in Table 1.4 (Chapter 1) at 100°F. Assume butanes to be all
© 2006 by Taylor and Francis Group, LLC
Compression 67
The C
VM
for the mixture is C
PM
− R 9.43 – 1.99 7.44 Btu/lb-mol-°R and
g 9.43/7.44 1.33.
The outlet temperature, T
2
, for a reversible adiabatic compression of an ideal
gas can be calculated by
T
2
T
1
(P
2
/P
1
)
( γ-1)/ γ
. (4.7)
However, the Engineering Data Book (2004a) recommends use of the following
empirical equation to obtain a more representative discharge temperature
(4.8)
where h
IS
is the entropic efficiency discussed below. In actuality, compressors
have some heat loss and are irreversible. The PV path followed in this case is
Pv
k
constant, (4.9)
where k is the empirically determined polytropic constant. This constant
replaces g in the above equations, an, thus, the polytropic work is
. (4.10)
Unless the value of is known, g is used for the typical quick calculations
on compressors.
All the components in natural gas behave like ideal gases at near-ambient
pressure and ambient and above temperatures. However, they become increasing
nonideal with increasing pressure. To estimate the work of compression, ideal
gas behavior can be assumed. However, to accurately determine the reversible
w
S
requires use of either an equation of state or thermodynamic charts, such as
a temperature–entropy or pressure–enthalpy diagram to accurately calculate h.
Example 4.2 Assume ideal gas behavior, and compute the reversible work required
to compress a natural gas mixture from 10 to 60 psig, both isothermally and
adiabatically, with an initial temperature of 80°F. Also, calculate the exit temperature
for adiabatic compression. The molar mass of the gas is 18, and the ratio of heat
capacities is 1.15. Assume the compressor is 100% efficient for this calculation.
Isothermal Compression
From Equation 4.2, the work is
w
S
−RT ln(P
2
/P
1
)/MW −1.986 (80 + 460) ln[(60 + 14.7)/(10 + 14.7)]/18
−65.9 Btu/lb
T T
P
P
2 1
2
1
1
1 1 +
|
.

`
}

]
]
]
]
|
.

`
}

− ( )/ γ γ

/ , η
IS
w
RT
MW
P
P
S



|
.

`
}

]
]
]
]

κ
κ
κ κ
1 2
1
1
1
1
( )
( )/
κ

© 2006 by Taylor and Francis Group, LLC
68 Fundamentals of Natural Gas Processing
Adiabatic Compression
Use of Equation 4.4 to compute the work, gives
The outlet temperature is given by Equation 4.7:
T
2
(80 + 460)[(60 + 14.7)/(10 + 14.7)]
(1.15−1)/ 1.15
624°R.
t
2
624 – 460 164°F
Thus, the adiabatic work of compression is about 8% higher than the isothermal
work.
4.2.2 MULTISTAGING
Multistaging of compressors is done for any of several reasons. An obvious reason
is that by cooling the gas between stages, the process reduces the gas volume,
and, thus, the work required. Although this reason is important, the major reason
for multistaging is materials limitations. Assuming g remained constant in Exam-
ple 4.1 and the outlet gas pressure was 625 psig, the outlet temperature would
be 365°F. Considering materials of construction, seals, and lubricants, the Engi-
neering Data Book (2004a) recommends about 300°F (150°C) as a “good aver-
age” for outlet temperatures. If high pressure is involved, 250 to 275°F (120 to
140°C) is recommended. Assuming the initial temperature is near ambient, these
maximum temperatures suggest pressure ratios of 3:1 to 5:1.
The minimum work is obtained when each stage of a multistage unit does
the same amount of work, and, thus, most compressors will have approximately
the same pressure ratio for each stage. In this case, the compression, or pressure
ratio, PR, for m stages is computed by
PR (P
2
/P
1
)
1/m
(4.11)
(Note that these compression ratios are always the ratio of absolute pressures.)
Then the total work of compression will be the sum of the work in each stage,
as computed by Equation 4.4.
Example 4.3 Determine the number of stages required to compress the gas in
Example 4.1 from 10 to 625 psig by use of a compression ratio of 3:1. Also, calculate
the exit temperature for each stage if the gas enters each stage at 80°F.
To determine the value of m, Equation 4.11 is rearranged to give
m ln(P
2
/P
1
)/ln (PR) ln[(625 + 14.7)/(10 + 14.7)]/ln (3) 2.96 or 3 stages.
The actual pressure ratio will be [(625 + 14.7)/(10 + 14.7)]
1/3
2.96. Equation 4.7
gives a discharge temperature of 161°F.
w
S

× × +


+ 1 15 1 986 80 460
18 1 15 1
1
60 . . ( )
( . )
114 7
10 14 7
70
1 15 1 1 15
.
.
( . )/ .
+
|
.

`
}

]
]
]
]


.. / 9Btu lb
© 2006 by Taylor and Francis Group, LLC
Compression 69
4.2.3 COMPRESSOR EFFICIENCIES
Several efficiencies are used to define compressor performance. The most com-
mon is the reversible adiabatic efficiency, commonly referred to as the isentropic
efficiency and defined as
h
IS
w
S
/w
S, ACTUAL
(4.12)
with w
S
computed by Equation 4.4.
The other commonly used efficiency, especially with centrifugal compressors,
is the polytropic efficiency, h
P
, defined as
h
P
[(g − 1)/g]/[(k – 1)/k]. (4.13)
This value approximates the ratio of isentropic work to the polytropic work
(i.e., Equation 4.4 divided by Equation 4.10), if the pressure-ratio term is ignored.
The polytropic efficiency will always be higher than the isentropic efficiency. It
can be used to calculate work when applied to Equation 4.10 and is useful for
estimating discharge conditions, as in Equation 4.14.
(4.14)
Example 4.4 Determine the isentropic efficiency of a compressor if the polytropic
efficiency is 77% and the gas has g 1.15. Assume the molar mass of the gas is
18, the inlet temperature is 80°F, and the compression ratio is 3.0.
Conversion from polytropic to adiabatic efficiency requires the following steps:
1. Calculate ideal polytropic work.
2. Calculate the actual work by dividing polytropic work by polytropic efficiency.
3. Calculate isentropic work.
4. Calculate isentropic efficiency from ratio of isentropic work to polytropic work.
Step 1. Both g and h
P
are given so Equation 4.13 can be used to determine k for use
in Equation 4.10. However, h
P
can be used in combination with g in Equation 4.10
to give
T T P P
p
2 1 2 1
1 1


( / )
( / )( )/ η γ γ
w
RT
MW
P
P
S
P
P



|
.

`
}


γ
γ
η γ γ
1 2
1
1 1
1
1
η
( )
( / )( )/

]
]
]
]

× + ) . 1.15 1.986(80
(1.15
460 0 77
18 −11)
(3.0) 72.1Bt
(1/0.77)(1.15-1)/1.15
1−

]
]
− uu/lb
© 2006 by Taylor and Francis Group, LLC
70 Fundamentals of Natural Gas Processing
Step 2. The actual work required is
w
S
−72.1/0.77 −93.6 Btu/lb
Step 3. The isentropic work required is
Step 4. The isentropic efficiency is then −70.3/−93.6 0.75or 75%, two percentage
points below the polytropic efficiency.
A volumetric efficiency also applies to reciprocating compressors. This effi-
ciency is the actual volume of gas in the piston cylinder divided by the piston-
displacement volume and includes the “dead” volumes associated with the valves
and other mechanical clearances, such as those between the piston and heads.
Considering only piston displacement, the volumetric efficiency, h
V
, is computed by
h
V
100 – (P
2
/P
1
) – C [(z
2
/z
1
) (P
2
/P
1
)
1/ γ
– 1], (4.15)
where z is the compressibility factor (z PV/RT), and C is the percent
clearance given by
C 100 (clearance volume)/(piston displacement volume) (4.16)
Ideally, vendor information on a compressor’s volumetric efficiency is avail-
able. However, once a compressor is overhauled, the data probably are inaccurate
(McCartney, 2005). Even if the compressor is several years old and has not been
overhauled, the data for that particular unit is important to have as manufacturers
modify efficiency data for newer models. The preferred way to obtain the current
volumetric efficiency on older units is to obtain a PV curve from a compressor
analyzer and back calculate the efficiency.
If valid volumetric efficiency information on the compressor is unavailable,
several corrections to the volumetric efficiency must be made (Engineering Data
Book, 2004a):
1. Subtract 4% for volumes around suction and discharge valving.
2. Subtract 5% for nonlubricated compressors to account for slippage
losses.
3. Subtract 4% for compressing propane or butane.
The first correction always applies, and if the last two apply, they are cumu-
lative. Therefore, the volumetric efficiency could be as much as 13% lower than
calculated by Equation 4.15.
w
S

× × +



1 15 1 986 80 460
18 1 15 1
1 3
1 15
. . ( )
( . )
( )
( . 11 1 15
70 3
)/ .
.

]
]
− Btu/lb
© 2006 by Taylor and Francis Group, LLC
Compression 71
The volumetric efficiency is required to estimate compressor capacity and
equivalent capacity. Section 4.4.1 discusses these calculations.
4.3 COMPRESSOR TYPES
Only reciprocating compressors were available when the natural gas industry
started. Although still used, they are being replaced by other compressor types.
In the 1980s, centrifugal compressors came into use. Rotary screw compressors
came into use in the 1990s. However, reciprocating compressors will remain
dominant in gas processing for many years to come. Table 4.2 (compiled from
numerous sources) lists the cost-effective ranges for various compressor types
used in gas plants, along with isentropic efficiency ranges.
inlet-gas flow rates at suction temperature and pressure (acfm) for the compressor
types discussed here. The entries serve as guidelines, as the values vary among
different sources. Other factors besides those listed in Table 4.2 must be considered
when the various compressor types are evaluated. These considerations are dis-
cussed below for the two compressor categories.
Compressors are driven by steam or gas-combustion turbines, electrical
motors, or internal-combustion engines running on fuel gas. Lower horsepower
TABLE 4.2
Typical Cost Effective Ranges of Compressors Used in Gas Processing
Maximum Pressure
psig (barg)
Inlet Flow Rate
a
acfm (m
3
/h) Inlet Discharge
Isentropic
Efficiency, %
Reciprocating
Single stage
Multistage
1 – 300
(2 – 500)
1 − 7,000
(2 – 12,000)
No limit
No limit
< 3,000 (200)
< 60,000
(4,000)
75 − 85
Centrifugal
Single stage
Multistage
50 – 3,000
(80 – 5,000)
500 – 200,000
(800 – 350,000)
No limit
No limit
1,500 (100)
10,000 (700)
70 − 75
Oil-free rotary
screw
< 40,000
(70,000)
< 150 (10) < 350 (25) 70 − 85
Oil-injected
rotary screw
< 10,000
(20,000)
< 400 (30) < 800 (60) 70 − 85
a
Compressor-gas volumes are based upon actual gas volumes at suction temperature and pressure.
Figure 4.1 shows the range of discharge pressures as a function of actual
© 2006 by Taylor and Francis Group, LLC
72 Fundamentals of Natural Gas Processing
and remote units tend to use electric motors. For high capacity applications, gas
turbines are often used.
4.3.1 POSITIVE DISPLACEMENT COMPRESSORS
A variety of positive displacement compressors are used. This section discusses
those that have a flow capacity of at least 10 acfm (actual cubic feet per minute)
(20 m
3
/h) and discharge pressures greater than 100 psig (7 barg). This excludes
diaphragm compressors, which have lower flow rates, and various rotary-vane
compressors, which have lower discharge pressures.
4.3.1.1 Reciprocating Compressors
Almost every onshore gas plant and field operation uses reciprocating compres-
being an established and known technology, they offer the highest efficiency, and
their performance is unaffected by gas composition. With well maintained ring
seals, they produce an oil-free output. Also, they can handle smaller gas volumes
and higher discharge pressures than other types of compressors. Only reciprocating
FIGURE 4.1 Ranges of discharge pressure as a function of actual flow rates for various
compressors. Two ranges exist for the reciprocating and centrifugal compressors. Lower
range denotes single-stage units and higher range denotes multiple-stage units. Legend:
— reciprocating, - - centrifugal, - ⋅ – ⋅ rotary screw, - ⋅ ⋅ – axial. (Adapted from Engineering
Data Book, 2004a.)
10
5
10
4
10
3
10
2 1 10
10
6
10
5
10
4
10
3
10
2
10
1
Inlet flow rate, acfm
D
i
s
c
h
a
r
g
e

p
r
e
s
s
u
r
e
,

p
s
i
g
sors. Figure 4.2 shows a reciprocating compressor on a gas line. In addition to
© 2006 by Taylor and Francis Group, LLC
Compression 73
and centrifugal compressors have sufficiently high discharge pressures to meet
the gas processing and gas-transmission requirements. Many reciprocating units
have been replaced by centrifugal compressors for high-volume service, but
reciprocating units remain as backups.
Reciprocating compressors are frequently multistaged and will have either
single-action or double-action pistons (Engineering Data Book, 2004a). Both
single-action and double-action pistons may be on a single frame.
Unfortunately, reciprocating compressors have a number of drawbacks; the
major one is reliability. Their runtime is 90 to 95% compared with about 99%
for other compressor types (Jandjel, 2000). However, units can have runtimes as
high as 99%. This reliability tends to be found in plants where operators can
watch the units and on electrically driven compressors, as gas-fired engines are
more prone to failure than the compressor (McCartney, 2005).
Their efficiency advantage decreases when the throughput is reduced. Remedies
for reduced volumes include
• Use of variable speed drives
• Adjustment of suction pressure
• Use of inlet-valve unloaders (which prevent the valves from completely
seating)
• Recirculation of gas to compensate for lower input rates
Also, the compression ratio typically is limited to about 3:1 to 4:1 for each
stage. A combination of high-temperature limitations and rod strength are the
constraining factors for the low ratio.
FIGURE 4.2 Reciprocating compressor at a gas wellhead. (Courtesy of Nuovo Pignone
Spa, Italy.)
© 2006 by Taylor and Francis Group, LLC
74 Fundamentals of Natural Gas Processing
Reciprocating compressors tend to be noisier than other types and are the
only ones requiring pulsed flow dampeners (snubbers). These compressors are
susceptible to condensed phases which erode cylinder walls, destroy lubricating
fluids and damage the rings. Finally, for the same volume of throughput, they
occupy much more space than the other types. In spite of the drawbacks, the
versatility of reciprocating compressors will keep them in the gas industry for a
long time.
Because of their widespread use, the Engineering Data Book (2004a) gives
extensive details on reciprocating compressors, including details on controls and
pulsation dampeners. The book also includes a troubleshooting guide.
4.3.1.2 Oil-Free Rotary Screw Compressors
Rotary screw compressors use two screws, or lobes (see Figure 4.3), to compress
the gas. Gas enters as the threads at the suction side are separating, and it moves
down the threads as the screws rotate. Clearances between the threads decrease and
compress the gas. The gas exits in an axial port at the end of the screws. A timing
drive keeps the two lobes synchronized. The screws run at 3,000 to 8,000 rpm, and
the speed is easily varied to provide an efficient means to handle lower flow rates.
Unlike reciprocating compressors, essentially all gas is displaced (i.e., the
volumetric efficiency is near 100%).
These compressors were first used in the steel industry to compress air
that contained up to 200-micron-sized particulates. Thus, they can easily
process streams containing condensate droplets. They are extremely reliable.
FIGURE 4.3 Cutaway view of an oil-injected rotary screw compressor. (Courtesy of Ariel
Corporation.)
© 2006 by Taylor and Francis Group, LLC
Compression 75
Jandjel (2000) states that condensed vapor with small g (e.g., isobutane) is
sometimes injected as a coolant that permits compression ratios to increase to
as high as 8:1.
The primary limitation of oil-free screw compressors is the low discharge
pressure. For gas processing, this limitation prevents these compressors from
being used for inlet and outlet compression. However, they are used for vapor
recovery and compression of low-pressure gases up to the suction pressure of
high-pressure compressors. The ability to inject liquid to increase the pressure
ratio makes them attractive for propane refrigeration units, although the oil-
injected compressors discussed next are widely used as well.
4.3.1.3 Oil-Injected Rotary Screw Compressors
enters in the upper right side, flows through the compressor, and exits at the lower
driven and the female screw follows. Unlike the oil-free rotary screw compressor,
the oil-injected type needs no timing drive. The injected oil provides the seal
between each screw and the casing and between the screws. Oil floods the unit
at various points along the screws. The oil-injected rotary screw compressor is
similar to the oil-free version but has two ports (Bruce, 2000). Like the oil-free
compressor, the oil-injected rotary screw compressor has an axial port, which in
this case, removes gas and remnants of oil. This compressor type also has a slide
valve located on the bottom of the compressor that removes the fluids through a
radial port.
The slide valve provides a major advantage for these compressors because
its position determines the volumetric throughput. The slide valve may be “fixed”
and thus require manual repositioning to adjust volumetric flows or may be
adjusted automatically. If the pressure ratio remains constant, some loss in effi-
ciency occurs when the volumetric throughput is adjusted because the optimum
volume ratio is a function of pressure ratio (Bruce, 2000). Bruce (2000) provides
details on turndown strategies for oil-injected screw compressors.
The oil system is a closed-loop system in these compressors and is a critical
component, as the oil provides both sealing and cooling. Discharged gas and oil go
to a scrubber that contains a demister to reduce oil concentration to less than
10 ppm (Bruce, 2000). The discharged oil is cooled from 170 – 220°F (75 – 105°C)
to 140 to 160°F (60 to 70°C) before being reinjected. Figure 4.3 shows tubing, which
is used to circulate the oil and two “bottles,” which are filters to remove particulates
from the oil. The tubing that goes to and from the extended piece on the lower right
drives the positioning of the slide valve on this particular unit.
Oil injection permits pressure ratios up to 23:1, but discharge pressures are still
below the highest pressures required for gas processing. These compressors are
well suited for and commonly used in propane refrigeration systems. Jandjel (2000)
estimates that these compressors will become the preferred compressors in the
future for vapor recovery and fuel gas compression. He cites the following reasons:
Figure 4.3 shows a cutaway view of an oil-injected rotary screw compressor. Gas
left. For oil-injected compressors, the male screw (upper one in Figure 4.3) is
© 2006 by Taylor and Francis Group, LLC
76 Fundamentals of Natural Gas Processing
• Improved lubricants that are both stable and inert
• Improved filtration systems that remove oil droplets down to less than
1 micron in size and to concentrations less than 1 ppm to meet oil-free
specifications
• Improved pressure and flow limits
These improvements address past concerns regarding these compressors.
However, oil compatibility with the process components can still be a problem
on new installations and selection of the proper oil is an empirical process.
Some newer units have extremely tight clearances between the lobes to
increase discharge pressure. A drawback to close tolerances is that if shut down,
the compressor must cool before being restarted.
For offshore applications, the oil-injected screw compressor has several
advantages over reciprocating compressors in low-pressure service, including:
• Small size, about one quarter of that required by reciprocating com-
pressor for comparable capacity
• Low maintenance
• Ease of adjusting to variable inlet flows
• High reliability
In addition to relatively low discharge pressure, the other disadvantages
involve the oil system. The lubricating oil must be compatible with the hydro-
carbons present at the operating conditions. Hydrocarbon solubility in the oil
must be minimal to prevent foaming problems on the suction side and to avoid
loss of the lubricating properties of the oil. Whereas oil-free screw compressors
handle “dirty gas” extremely well, oil-injected systems are more susceptible to
solids. If solids are a problem, the oil filtration systems must be maintained
properly. Energy is spent to compress the oil, but this is work comparable to
pumping liquid and is trivial compared with the work of gas compression.
4.3.2 DYNAMIC COMPRESSORS
The two dynamic compressor types are centrifugal and axial. Axial compressors
handle large gas volumes (50 to 300 Macfm [80 to 500 Mam
3
/h]) and have higher
efficiencies than do centrifugals. However, they generally have discharge pres-
sures below 200 psig (14 barg) and are restricted to clean gases. Although not
used for processing the natural gas, axial compressors will be discussed briefly
in the context of their use in the gas turbines that drive centrifugal compressors.
shows a large gas turbine compressor system. The centrifugal compressor is the
large cylindrical unit in the center of the picture, with the two large attached
pipes. The large ductwork to the left of the compressor is for the turbine exhaust
gas, and the one further to the left is intake air to the axial turbine.
To illustrate how axial and centrifugal compressors work in tandem, Figure 4.4
shows a schematic of a typical gas turbine−driven centrifugal compressor. Figure 4.5
© 2006 by Taylor and Francis Group, LLC
Compression 77
4.3.2.1 Centrifugal Compressors
Since the 1980s, centrifugal compressors have been replacing reciprocating com-
pressors and are widely used throughout the industry. With their high flow rates
and relatively high discharge pressures, they commonly provide both inlet and
outlet compression. They also are used extensively for gas pipeline transmission.
FIGURE 4.4 Schematic of a gas turbine-driven centrifugal compressor.
FIGURE 4.5 Large gas turbine-driven centrifugal compressor. (Courtesy of Solar Tur-
bines.)
Low pressure gas
High
pressure gas
Fuel gas
Clean
air
Axial
compressor
Combustion
turbine
Exhaust gas
Centrifugal
compressor
© 2006 by Taylor and Francis Group, LLC
78 Fundamentals of Natural Gas Processing
In gas processing, they are usually powered by either a gas turbine or a turboex-
pander but can be steam or electrically driven.
Figure 4.6 shows a cutaway view of a multistage centrifugal compressor. Basi-
cally, the compressor consists of a rapidly rotating (1,000 to 20,000 rpm for natural
gas applications) impeller (rotor), surrounded by a stationary diaphragm. (For more
details on these compressors, see the Engineering Data Book [2004a].) Centrifugal
compressors work by having gas enter the rotor of each stage near the rotor shaft.
The gas is thrust outwards into the channels of the stationary diaphragm, which
are called the diffuser or diffuser passages, where the velocity converts to pressure.
The diffuser contains vanes to maximize conversion of velocity to pressure and to
direct gas into the next stage. Vendors estimate that about one third of the pressure
increase occurs in the diffuser, and the rest comes from the rotors.
The rotor can be open or shielded on one or both sides. A survey of literature
from compressor manufacturers indicates that most compressors for natural gas
service are shielded on both sides (i.e., closed), as shown in Figure 4.6. The
blades in the rotor are swept back to improve aerodynamics and mechanical
stability. A nice feature of these compressors is that all stages are on the same
drive shaft.
Like other compressors, exit temperature limits the pressure ratio in centrifugal
compressors. The Engineering Data Book (2004a) recommends a maximum
discharge temperature of 400°F (200°C).
A major difference between positive displacement and centrifugal compressors
FIGURE 4.6 Cutaway of a centrifugal compressor. (Courtesy of Solar Turbines.)
is how discharge pressure affects gas flow. Figure 4.7 shows the interaction
© 2006 by Taylor and Francis Group, LLC
Compression 79
between head (i.e., discharge pressure) and actual flow rate for reciprocating,
axial, and variable speed centrifugal compressors. For reciprocating compressors,
the gas flow rate is constant and independent of the downstream pressure (constant
volume – variable head), which is why relief valves or other devices are necessary
to protect the compressor from excessive discharge pressures.
For centrifugal compressors, the downstream pressure greatly affects the
throughput (constant head – variable volume). The range over which the com-
pressor functions properly is limited. If downstream pressure becomes too high
(or inlet pressure becomes too low), the compressor cannot push the gas out and
backflow occurs, which results in a drop in the discharge pressure. Then the
compressor raises the discharge pressure only to go into backflow again. This
cyclic process is called “surge” and must be avoided because the pressure oscil-
lations cause overheating and vibration that can damage internal bearings. All
centrifugal compressors come with surge control systems that protect the com-
pressor by gas recirculation. The Engineering Data Book (2004a) discusses surge-
control methods.
The other limit occurs when the gas flow through the compressor reaches
sonic velocity and flow cannot be increased. This limit is choked flow, or “stone-
wall.” To increase the flow requires internal modifications.
Centrifugal compressors have other disadvantages. Perhaps the most impor-
tant is the effect of gas density on discharge pressure. This effect is not a factor
where the gas composition remains nearly constant, such as in pipeline transmission
boosters. However, if used in applications where gas composition changes over
time, the compressors must be restaged periodically to obtain maximum efficiency
FIGURE 4.7 Relationship between compressor and inlet-volume flow rate for centrifugal,
axial, and reciprocating compressors. (Adapted from Engineering Data Book, 2004a.)
13-27
Inlet volume flow rate
P
r
e
s
s
u
r
e

h
e
a
d
S
u
r
g
e

l
i
n
e
S
t
o
n
e
w
a
l
l

l
i
n
e
Centrifugal
Reciprocating
Axial
© 2006 by Taylor and Francis Group, LLC
80 Fundamentals of Natural Gas Processing
while maintaining a constant flow rate and discharge pressure. Restaging is
expensive in terms of both capital outlay and downtime.
Gases that flow into centrifugal and axial compressors must be free of con-
densed phases because the particles damage the vanes. Their off-design operation
is poor compared with the other compressors, and they are not good for inter-
mittent operations. However, where they can be used, their reliability readily
offsets their drawbacks.
4.3.2.2 Axial Flow Compressors
Axial compressors routinely compress large volumes of gas (50 to 500 Macfm
[80 to 800 Mam
3
/h]) at pressures up to about 200 psig (14 barg), although in
special applications, they may go to higher pressures. They are smaller and more
efficient than centrifugals. As the name implies, axial compressors use stationary
and rotating vanes to push the gas down the axis instead of in a radial direction
like centrifugal compressors.
compressors, but in combination with a gas turbine and centrifugal compressor
they operate in a range where neither surge nor stonewalling occurs. They have
a higher efficiency than centrifugals and are ideal for moving large volumes of
air, such as that needed to fire the gas turbines used to drive compressors and
generators. The remainder of this section focuses on axial compressors combined
with gas turbines.
A gas-fired turbine works on the same principle as a jet engine. Figure 4.8
shows a cutaway view of an axial compressor in combination with a gas turbine.
Clean air enters the axial compressor at the lower left side of the drawing. It is
compressed in the 11 stages of the axial compressor to an exit temperature of
650 to 950°F (350 to 500°C). The hot air burns the fuel in the combustion zone
FIGURE 4.8 Cutaway view of a gas-fired turbine. (Courtesy of Nuovo Pignone Spa, Italy.)
Figure 4.7 shows that axial compressors have a similar behavior to centrifugal
© 2006 by Taylor and Francis Group, LLC
Compression 81
and increases the gas temperature to 1,800 to 2,300°F (1,000 to 1,250°C). The
high-temperature, high-pressure gas then passes over the three turbine blades to
generate the mechanical energy and exits at approximately 950°F (500°C).
Like all heat engines, the turbine thermal efficiency is driven by temperature
differences. Gas turbines follow the Brayton cycle (Smith et al., 2001), and ideal
efficiencies (defined as the ratio of net work generated divided by heat input) can
be around 40%. However, company literature indicates actual overall efficiencies
in the 28 to 35% range. Turbine ratings are often expressed in terms of heat rate,
which is the ratio of the heat generated to the power produced (Btu/hp-h or kJ/kW-
h). Note that the lower heating value is used in this calculation. The Engineering
Data Book (2004b) provides additional details on the calculation of heat rates and
power on the basis of operating conditions. The book also describes combined-
cycle units that use the waste heat to generate steam for additional power generation.
Example 4.5 Compute the overall thermal efficiency of a gas turbine that has a
heat rate of 8,500 Btu/hp-h (12,000 kJ/kWh).
The thermal efficiency is the reciprocal of the heat rate. Including the conversion
between energy and power (2,544 Btu/hp-h or 3,600 kJ/kW-h), the thermal efficiency is
Thermal Efficiency 2,544/heat rate (3,600/heat rate)
2,544/8,500 (3,600/12,000) 0.30 or 30%
(If the heat rate is stated in Btu/kWh, the conversion is 3,414 Btu/kW-h.)
For applications in which the speed is nearly constant, all rotating components
are on a single shaft. However, if a variable-speed centrifugal compressor is
needed, a second shaft will be coupled to the turbine shaft to permit the com-
pressor and turbine to run at different speeds.
The axial compressor–gas turbine unit is known for high reliability. However,
the greater number of moving parts and seals, along with high temperature operation,
makes the turbine the least reliable of the turbine–centrifugal compressor combina-
tion. Air filtration systems are mandatory to protect the axial compressor from
particulate matter. In offshore applications, demisters replace the filters. Self-cleaning
filters that use pulses of clean air from one filter bank to blowout another filter bank
are used in many environments. They have the advantage of being able to de-ice a
filter bank at the same time when in cold-weather service. The Engineering Data
Book (2004b) provides additional recommendations on filtration systems.
Plant operators watch the combustion zone exhaust temperature closely
because it runs close to materials limits. High ambient temperatures sometimes
raise the exhaust temperature to a point where gas throughput must be curtailed.
4.4 CAPACITY AND POWER CALCULATIONS
This section provides some methods for estimating the capacity and power
requirements of compressors when ideal gases or nearly ideal gases are used.
The Engineering Data Book (2004a) provides more detailed calculations. Most
© 2006 by Taylor and Francis Group, LLC
82 Fundamentals of Natural Gas Processing
of the calculations are independent of compressor type. Screw compressors with
liquids present require either polytropic or more rigorous thermodynamic methods
to be used. Reciprocating compressors require knowledge of displacement
volumes to compute flow rates, as will be discussed below. If accurate calculations
are needed, equation of state methods should be used, along with information
about the compressor performance provided by the vendor.
4.4.1 CAPACITY
4.4.1.1 General Calculations
In this section, we assume that the gas can be described by
PV nzRT, (4.17)
Example 4.6 Estimate the compressibility factor, z, for the gas composition given
in Table 4.3 at 120°F and 600 psig.
One commonly used method for estimating the compressibility factor for a gas or
assumes that the compressibility factor for any gas with a molecular weight less
than 40 is a function of the reduced pressure, P
R
, and temperature, T
R
, of the gas.
(For more details on this “corresponding states” method, see a book on physical
properties [e.g., Poling et al., 2000]). For a pure gas, the reduced pressure and
temperature is obtained by division of the absolute temperature by the critical
temperature and the absolute pressure by the critical pressure. For mixtures, the
critical properties are obtained by use of pseudocritical pressure and temperature
for the mixture defined by (Engineering Data Book, 2004c)
TABLE 4.3
Data for Example 4.6
Component Mol fraction, y P
c
, psia yP
c
T
c
, °R yT
c
Methane 0.88 666 586.1 343 301.8
Ethane 0.06 707 42.4 550 33.0
Propane 0.03 616 18.5 666 20.0
i-butane 0.01 528 5.3 734 7.3
Butane 0.02 551 11.0 765 15.3
Mixture properties 663.3 377.5
P x P T x T
CM i C CM i C
i i

∑ ∑
and .
where z is the compressibility factor. (Appendix B provides charts to estimate z.)
gas mixture is to use a general compressibility factor chart (see Appendix B), which
© 2006 by Taylor and Francis Group, LLC
Compression 83
The reduced temperature for the mixture is (120 + 460)/377.5 1.54, and
the reduced pressure is (600 + 14.7)/663.3 0.93. From the compressibility chart
For capacity calculations, the actual volumetric flow rate, Q, is based upon
inlet conditions to each stage. To compute the actual flow rate from gas flows
given as standard flow rates use
(4.18a)
, (4.18b)
where the subscript “1” denotes inlet conditions. The subscript “R” refers to
the conditions used to define the reference state, in this case, 14.7 psia (1.01 bara)
and 60°F (15.6°C). In these equations, the units must be consistent and the
temperatures are absolute (°R or K). If mass flow rates are given, the molar mass
(MW), of the gas as well as the mass flow rate, must be known:
(4.19a)
(4.19b)
If the molar flow rate , is given, the actual gas flow rate is computed by
(4.20a)
(4.20b)
4.4.1.2 Reciprocating Compressors
Calculating the throughput of a reciprocating compressor requires knowledge of
the volume displaced during each stroke. Normally, the piston displacement, PD,
is in ft
3
/min (m
3
/h), which is the product of cross-sectional area, stroke travel,
and RPM. The equations for computing PD depend upon compressor type and
are given below, in dimensioned form (Engineering Data Book, 2004a).
Q scfm
P psia
T R z

|
.

`
}

°
|
.

`
}

14 7
520
1
1 1
.
( )
( )
zz
R
|
.

`
}

Q sm h
P bar
T K z
z
R

|
.

`
}

|
.

`
}

|
.

3
1
1 1
101
288 ( )
( )
``
}

w,
Q
w lb T R
MWP psia
z
z
R

°
|
.

`
}
10 73
1
1
1
. ( / min) ( )
( )

Q
w kg hr T K
MWP Bar
z
z
R

|
.

`
}

8 314
1
1
1
. ( / ) ( )
( )

n
Q n lb mol hr
T R
P p

°
|
.

`
}

379 5
520
14 7
1
1
. ( / )
( )
.
(

ssia
z
z
R
)
|
.

`
}

|
.

`
}

1
Q n mole hr
T K
P Bar
z
z
R

|
.

`
}

|
.

8 314
1
1
1
. ( / )
( )
( )

``
}

in Appendix B, the compressibility factor is 0.91.
© 2006 by Taylor and Francis Group, LLC
84 Fundamentals of Natural Gas Processing
For a single-acting piston compressing only on the outer end (familiar
configuration):
(4.21a)
(4.21b)
For a single-acting piston compressing only on the crank end:
(4.22a)
(4.22b)
For double-acting piston compressing on both the crank and piston end
(excluding the tail rod type, which has rods on both side of the piston):
(4.23a)
(4.23b)
With the displacement volume, suction pressure and volumetric efficiency
(Equation 4.15), the net capacity can be computed. Note that compressor vendors
PD
D in
RPM
rev
Stroke in
ft
×
|
.

`
}

× ×
π
2 2 3
4
( )
min
( )
11728
4 55 10
3
4 2
in
D RPM Stroke × × ×

. ( )
PD
D cm
RPM
rev
Stroke cm
m
×
|
.

`
}

× ×
π
2 2 3
4 1
( )
min
( )
00
60
4 7 10
6 3
5 2
cm h
D vRPM Stroke
×
× ×

min
. ( )
PD
D d in
RPM
rev
Stroke in

×
|
.

`
}

×
π( )( )
min
(
2 2 2
4
))
. ( ) ( )
×
× − × ×

ft
in
D d RPM Stroke
3
3
4 2 2
1728
4 55 10
PD
D d cm
RPM
rev
Stroke cm

×
|
.

`
}

×
π( )( )
min
(
2 2 2
4
))
min
. ( ) ( )
× ×
× − × ×

m
cm h
D d RPM St
3
6 3
5 2 2
10
60
4 7 10 rroke
PD
D d in
RPM
rev
Stroke i

×
|
.

`
}

×
π( )( )
min
(
2
4
2 2 2
nn
ft
in
D d RPM Stro
)
. ( ) ( )
×
× − × ×

3
3
4 2 2
1728
4 55 10 2 kke
PD
D d cm
RPM
rev
Stroke c

×
|
.

`
}

×
π( )( )
min
(
2
4
2 2 2
mm
m
cm h
D d RPM
)
min
. ( ) ( )
× ×
× − × ×

3
6 3
5 2 2
10
60
4 7 10 2 SStroke
© 2006 by Taylor and Francis Group, LLC
Compression 85
usually use a standard pressure of 14.4 psia* (0.99 bar), instead of 14.7 psia. In
dimensioned form, the equivalent volume is
(4.24a)
(4.24b)
4.4.2 POWER REQUIREMENTS
The ideal power requirements for isentropic compression makes use of the shaft
work calculated in Equation 4.4 and the isentropic efficiency to give
Power z
avg
w
S
, (4.25)
where is the mass flow rate, and z
avg
is the average of the inlet and outlet
compressibility factors.
Because they are the more common compressor types, the Engineering Data
Book (2004a) gives some useful methods for estimating power requirements for
reciprocating and centrifugal compressors. The following paragraphs give these
relationships.
4.4.2.1 Reciprocating Compressors
An empirical equation (Engineering Data Book, 2004a) for estimating the power
required for reciprocating compressors is
Brake HP 22 F PR m MMacfd (4.26a)
Brake HP 0.014 F PR am
3
/h, (4.26b)
* The 14.4 psia value was selected long ago as a matter of convenience to make calculations simpler
because a day contains 1,440 minutes; see, for example, Equation 4.24a (McCartney, 2005).
MMscfd
z
z
PD
ft
P
V
×
|
.

`
}

× ×
14 4
1
3
1
100 14 4
1
.
min .
η
4440 10
10
6 3
3
1
1
6
min
d
MMft
ft
z
PD P
V
×
× × ×


η
, assum ming
14.4
Z
Sm d
z
z
PD
m
h
P
h
d
bar V
3
1
1
3
1
100 1
24
0 24
×
|
.

`
}

× ×

η
. ×× × ×
η
V
z
PD P
1
1
, assuming z at 1 bar = 1.
m
m
© 2006 by Taylor and Francis Group, LLC
86 Fundamentals of Natural Gas Processing
where Brake HP is the actual work delivered to the compressor and the
correction factor F is given by
F 1.0 for single-stage compression (m 1)
F 1.08 for two-stage compression (m 2)
F 1.10 for three-stage compression (m 3)
The volume should be based upon inlet temperature and 14.4 psia (1.0 bara).
One should be aware of several items when doing these calculations. First,
vendors rate compressors on the basis of 14.4 instead of 14.7 psia. The above
equation was developed for large low-speed (300 to 400 rpm) compressors that
process gases with a specific gravity of 0.65 and PR values greater than 2.5. For
gases with specific gravities of 0.8 to 1.0, use 20 in Equation 4.26a or 0.013 in
Equation 4.26b; the value should be 16 to 18 (10 to 12) for PR in the 1.2 to 2 range.
Example 4.7 Use Equation 4.26 to estimate the brake horsepower needed to compress
30 MMscfd of the gas in Example 4.1 from 10 to 625 psig. Assume the intake
temperature to be 80°F.
Example 4.3 showed that three stages are necessary to go from 10 to 625 psig; therefore,
the factor F is 1.10. Also, the example found the compression ratio, PR, to be 2.96.
The molar mass of the gas is 18, so the specific gravity is 18/28.96 0.62. Therefore,
the factor of 22 is appropriate. Finally, use Equation 4.18, and assume the compress-
ibility factors to be unity, to correct the volumetric flow rate to 14.4 psia and 80°F:
Then the estimated horsepower requirement is
Brake HP 22 F PR m MMcfd 22 × 1.10 × 2.96 × 3 × 31.8 6,830 hp (5.1 MW).
4.4.2.2 Centrifugal Compressors
As in the case of centrifugal pumps, head, h
d
(the height of a column of the
fluid being pumped, [ft-lb
f
/ lb
m
, N
.
m/kg]) is commonly used in work with
centrifugal compressors, instead of pressure. The isentropic head, h
dIS
, is given
by
, (4.27)
MMcfd MMscfd
P psia
T R

|
.

`
}

°
|
.

`
14 7
520
1
1
.
( )
( )
}}

|
.

`
}

+
|
.

`
}

30
14 7
14 4
460 80
520
31 8
.
.
. MMMcfd
h z w
z
MW
RT P
P
dIS Avg S
Avg


|
.

`
}


γ
γ
γ γ
1 2
1
1
1
( )/
−−

]
]
]
]
1
© 2006 by Taylor and Francis Group, LLC
Compression 87
where z
avg
is the average compressibility factor between inlet and discharge
temperature and pressure. If polytropic head, h
P
, is needed, replace g with k. The
relationship between isentropic and polytropic head is
h
dP
/h
dIS
h
p
/h
IS
. (4.28)
The actual power in compressing the gas is
Power h
dIS
/h
IS
(4.29)
for isentropic compression. The isentropic power can be calculated by use of
the isentropic head and efficiency. For an estimation of the mechanical losses,
the Engineering Data Book (2004a) suggests
Mechanical losses (Power)
0.4
(4.30)
so that the brake HP requirement will be the sum of Equation 4.29 and
Equation 4.30.
To provide variable flow rates, centrifugal compressors vary the speed of the
compressor. The effect of RPM on compressor performance follows the affinity,
or fan laws, which are valid for single-stage compressor or multistage compressor
with low compression ratios or very low Mach numbers. The fan laws give
(4.31)
(4.32)
. (4.33)
Although these relations are useful, they become less accurate with deviations
of more than about ±10% from design conditions.
4.5 COMPARISON OF RECIPROCATING AND
CENTRIFUGAL COMPRESSORS
Only reciprocating and centrifugal compressors can meet the demand for dis-
charge pressures in the 1,000 psi (70 bar) range. When gas volumes are large
enough for centrifugal compressors to be considered, they have many advantages
over reciprocating compressors, including (Engineering Data Book, 2004a):
m
Q
Q
RPM
RPM
1
2
1
2

h
h
RPM
RPM
1
2
1
2
2

|
.

`
}

BrakeHP
BrakeHP
RPM
RPM
1
2
1
2
3

|
.

`
}

© 2006 by Taylor and Francis Group, LLC
88 Fundamentals of Natural Gas Processing
• Lower initial and maintenance costs
• Higher reliability and longer intervals between downtimes
• Less operating attention
• Greater throughput per unit of platform area
• Adaptability to high speed, low-maintenance drivers
Both efficiency and maintenance must be considered when reciprocating and
centrifugal compressors are compared. Reciprocating compressors are far more effi-
combustion engines widely vary with power output and vendor, which makes a
general comparison meaningless. (The Engineering Data Book [2004b] lists these
data for major vendors.) If the gas processor shares the fuel costs with the producer
factor for the gas processor, because most of the fuel cost is borne by the producer.
However, the processor typically pays all maintenance costs. These costs for
reciprocating compressors run $40 to $60/hp/year ($30 to $45/kW/yr). For cen-
trifugal compressors, the cost is $10 to $15/hp/yr ($8 to $12/kW/yr) (McCartney,
2005). If a 100 MMscfd (2.8 Mm
3
/d) plant has 70,000 hp (50 MW) devoted to
inlet and outlet compression, the savings by use of centrifugal compressors can
be several million dollars per year.
4.6 SAFETY AND ENVIRONMENTAL
CONSIDERATIONS
Apart from the dangers inherent in high-pressure equipment, noise is the main
safety concern related to compression, especially for compressors on the scale
of centrifugal units. Normally ear protection is worn in the plant, but it is a must
around compressors, which produce noise levels in the 85 db to 95 db range.
Exhaust emissions from compressor power drivers are the major environmental
concern. Newer internal-combustion engines that drive reciprocating compressors
were designed to meet an emission level of 2 g NO
X
and CO per horsepower by
use of “clean burn or lean burn” units (Kuchinski, 2005). However, tighter controls
are being imposed and catalytic converters, which reduce NO
X
to N
2
and oxidize
CO to CO
2
, are being installed on these units. This requirement applies to smaller
booster station compressors as well as large-plant units. This requirement is one
of the reasons that smaller booster stations use electric motor−driven compressors
where reliable power is available.
High air-to-fuel ratios combined with good combustor designs minimize CO
emission from gas turbines. Nitrogen oxides are the most significant exhaust pollutant
and high-efficiency systems enhance NO
X
production. Water is often added to reduce
NO
X
production. This addition helps reduce the NO
X
formed from N
2
but not that
formed from organic nitrogen, which might be in the fuel. Water may actually enhance
NO
X
production from the organic nitrogen, as well as increase CO production
(Engineering Data Book, 2004b). Modification of the air-fuel ratio and combustor
design helps keep the NO
X
exit concentration level in the 10 to 15 ppm range.
cient (see Table 4.2) on the compressor side. (Heat rates of gas turbines and internal-
(e.g., with POP contracts [see Chapter 2]), the difference in efficiencies is not a major
© 2006 by Taylor and Francis Group, LLC
Compression 89
REFERENCES
Bruce, J. T., Screw compressors: A Comparison of Applications and Features to Conven-
tional Types of Machines, Proceedings of the Seventy-Ninth Annual Convention
of the Gas Processors Association, Tulsa, OK, 2000, 229.
Crum, F.S., Application of J-T Plants for LP-Gas Recovery, Proceedings of the Sixtieth
Annual Convention of the Gas Processors Association, Tulsa, OK, 1981, 68.
Engineering Data Book, 12th ed., Sec. 13, Compressors and Expanders, Gas Processors
Supply Association, Tulsa, OK, 2004a.
Engineering Data Book, Prime Movers for Mechanical Drives, Gas Processors Supply
Association, Tulsa OK, 2004b.
Engineering Data Book, 12th ed., Sec. 23, Physical Properties, Gas Processors Supply
Association, Tulsa OK, 2004c.
Jandjel, D.G., Select the right compressor, Chem. Eng. Progress, 96 (7) 15, 2000.
Kuchinski, J., private communication, 2005.
McCartney, D.G., private communication, 2005.
Poling, B.E., Prausnitz, J.M., and O’Connell, J.P., The Properties of Gases and Liquids,
5th ed., McGraw Hill, New York, 2000.
Smith, J.M., Van Ness, H.C., and Abbott, M.M., Introduction to Chemical Engineering
Thermodynamics, 6th ed., McGraw-Hill, New York, 2001.
© 2006 by Taylor and Francis Group, LLC
91
5
Gas Treating
5.1 INTRODUCTION
Gas treating involves reduction of the “acid gases” carbon dioxide (CO
2
) and
hydrogen sulfide (H
2
S), along with other sulfur species, to sufficiently low levels
to meet contractual specifications or permit additional processing in the plant
without corrosion and plugging problems. This chapter focuses on acid gases
Field Liquids Removal
Gas & Liquids from Wells
Inlet
Compression
Sulfur
Recovery
Gas Treating
Dehydration
Nitrogen
Rejection
Helium
Recovery
Outlet
Compression
Elemental
Sulfur
Field Dehydration
Field Acid Gas Removal
Field Compression
Sales Gas
Water &
Solids
Natural
Gas Liquids
Natural
Gasoline
Raw Helium
CO
2
Liquefaction
Liquefied
Natural Gas
Inlet Receiving
Hydrocarbon
Recovery
Liquids Processing
© 2006 by Taylor and Francis Group, LLC
92 Fundamentals of Natural Gas Processing
because they are the most prevalent. When applicable, discussion is given to other
sulfur species.
In this introduction, we briefly address the following questions to set the
stage for the remainder of the chapter:
1. Why are the acid gases a problem?
2. What are the acid gas concentrations in natural gas?
3. How much purification is needed?
4. What is done with the acid gases after separation from the natural gas?
5. What processes are available for acid gas removal?
5.1.1 THE PROBLEM
Hydrogen sulfide is highly toxic, and in the presence of water it forms a weak,
corrosive acid. The threshold limit value (TLV) for prolonged exposure is 10 ppmv
and at concentrations greater than 1,000 ppmv, death occurs in minutes (Engi-
neering Data Book, 2004b). It is readily detectable at low concentrations by its
“rotten egg” odor. Unfortunately, at toxic levels, it is odorless because it deadens
nerve endings in the nose in a matter of seconds.
When H
2
S concentrations are well above the ppmv level, other sulfur species
can be present. These compounds include carbon disulfide (CS
2
), mercaptans
(RSH), and sulfides (RSR), in addition to elemental sulfur. If CO
2
is present as
well, the gas may contain trace amounts of carbonyl sulfide (COS). The major
source of COS typically is formation during regeneration of molecular-sieve beds
Carbon dioxide is nonflammable and, consequently, large quantities are unde-
sirable in a fuel. Like H
2
S, it forms a weak, corrosive acid in the presence of water.
The presence of H
2
S in liquids is usually detected by use of the copper
strip test (ASTM D1838 Standard test method for copper strip corrosion by
liquefied petroleum (LP) gases). This test detects the presence of materials that
could corrode copper fittings. One common method of determining ppm levels
of H
2
S in gases is to use stain tubes, which involves gas sampling into a glass
tube that changes color on the basis of H
2
S concentration. This method is good
for spot and field testing. A newer, continuous method measures reflectance off
of a lead acetate−coated tape that darkens in the presence of H
2
S. The test
(ASTM D4084 Standard test method for analysis of hydrogen sulfide in gaseous
fuels (Lead Acetate reaction rate method) 2005) detects H
2
S down to 0.1 ppmv.
It is specific to H
2
S but used to detect total sulfur by hydrogenating the sulphur
species in the gas before going to the detector.
5.1.2 ACID GAS CONCENTRATIONS IN NATURAL GAS
Although many natural gases are free of objectionable amounts of H
2
S and
CO
2
, substantial quantities of these impurities are found in both gas reserves
used in dehydration (see Chapter 6).
© 2006 by Taylor and Francis Group, LLC
Gas Treating 93
and production in the United States. In a survey of U.S. gas resources, Meyer
(2000) defined subquality gas as that containing CO
2
≥ 2%, N
2
≥ 4%, or H
2
S
≥ 4 ppmv. These criteria were selected because gases that contain these
amounts of impurities generally require upgrading or blending. Using these
criteria, Meyer (2000) estimated that 41% of proven gas reserves are subquality
and 34% of 1996 gas production in the lower 48 states was subquality. For
associated gas reserves, they estimated that 13% was subquality. It is obvious
that removal of H
2
S and CO
2
is a major concern in gas processing.
5.1.3 PURIFICATION LEVELS
The inlet conditions at a gas processing plant are generally temperatures near
ambient and pressures in the range of 300 to 1,000 psi (20 to 70 bar), so the
partial pressures of the entering acid gases can be quite high. If the gas is to
be purified to a level suitable for transportation in a pipeline and used as a
residential or industrial fuel, then the H
2
S concentration must be reduced to
0.25 gr/100 scf (6 mg/m
3
) (Engineering Data Book, 2004a), and the CO
2
concentration must be reduced to a maximum of 3 to 4 mol%. However, if the
gas is to be processed for NGL recovery or nitrogen rejection in a cryogenic
turboexpander process, CO
2
may have to be removed to prevent formation of
solids. If the gas is being fed to an LNG liquefaction facility, then the maximum
CO
2
level is about 50 ppmv (Klinkenbijl et al., 1999) because of potential
solids formation.
5.1.4 ACID GAS DISPOSAL
What becomes of the CO
2
and H
2
S after their separation from the natural gas?
The answer depends to a large extent on the quantity of the acid gases.
For CO
2
, if the quantities are large, it is sometimes used as an injection
fluid in EOR (enhanced oil recovery) projects. Several gas plants exist to
support CO
2
flooding projects; the natural gas and NGL are valuable by-
products. If this option is unavailable, then the gas can be vented, provided it
satisfies environmental regulations for impurities. Moritis (2001) gives a gen-
eral discussion of the state of EOR in the United States. Although the United
States has not ratified the Kyoto Treaty, which limits CO
2
emissions to the
atmosphere, caps may be placed on these emissions because of the prepon-
derance of scientific data that shows CO
2
is a significant contributor to global
warming.
In the case of H
2
S, four disposal options are available:
1. Incineration and venting, if environmental regulations regarding
sulfur dioxide emissions can be satisfied
2. Reaction with H
2
S scavengers, such as iron sponge
© 2006 by Taylor and Francis Group, LLC
94 Fundamentals of Natural Gas Processing
3. Conversion to elemental sulfur by use of the Claus or similar process
4. Disposal by injection into a suitable underground formation, as discussed
by Wichert and Royan (1997) and Kopperson et al. (1998 a,b)
The first two options are applicable to trace levels of H
2
S in the gas, and the last
two are required if concentrations are too high to make the first two options feasible.
5.1.5 PURIFICATION PROCESSES
Four scenarios are possible for acid gas removal from natural gas:
1. CO
2
removal from a gas that contains no H
2
S
2. H
2
S removal from a gas that contains no CO
2
3. Simultaneous removal of both CO
2
and H
2
S
4. Selective removal of H
2
S from a gas that contains both CO
2
and H
2
S
Because the concentrations of CO
2
and H
2
S in the raw gas to be processed
and the allowable acid gas levels in the final product vary substantially, no
single process is markedly superior in all circumstances, and, consequently,
many processes are presently in use. Table 5.1 summarizes the more important
TABLE 5.1
Acid Gas Removal Processes

Details of processes are available in Kohl and Nielson (1997).
(see Chapter 11)
© 2006 by Taylor and Francis Group, LLC
Gas Treating 95
processes and groups them into the generally accepted categories. Details of
processes not discussed in this chapter may be found in the handbook by Kohl
and Nielsen (1997).
Some of the more important items that must be considered before a process
is selected are summarized from the Engineering Data Book (2004b):
• The type and concentration of impurities and hydrocarbon composition
of the sour gas. For example, COS, CS
2
, and mercaptans can affect the
design of both gas and liquid treating facilities. Physical solvents tend
to dissolve heavier hydrocarbons, and the presence of these heavier
compounds in significant quantities tends to favor the selection of a
chemical solvent.
• The temperature and pressure at which the sour gas is available. High
partial pressures (50 psi [3.4 bar] or higher) of the acid gases in the
feed favor physical solvents, whereas low partial pressures favor the
amines.
• The specifications of the outlet gas (low outlet specifications favor the
amines).
• The volume of gas to be processed.
• The specifications for the residue gas, the acid gas, and liquid products.
• The selectivity required for the acid gas removal.
• The capital, operating, and royalty costs for the process.
• The environmental constraints, including air pollution regulations and
disposal of byproducts considered hazardous chemicals.
If gas sweetening is required offshore, both size and weight are additional
factors that must be considered. Whereas CO
2
removal is performed offshore,
H
2
S removal is rarely done unless absolutely necessary because of the problems
of handling the rich acid gas stream or elemental sulfur.
Selection criteria for the solvent-based processes are discussed by Tennyson
dations. The guidelines in these figures are naturally approximate and should
be treated as such. These figures are for solvent-based processes only. Thus,
they exclude some commonly used processes such as adsorption and mem-
branes. We slightly modified their recommendations to make it more current,
but these guidelines, established in 1977, still are useful. Note that “hybrid” in
the figures denotes mixed-solvent systems that contain both amine and a phys-
ical solvent.
the more widely used acid gas processes, including degree of purification attainable,
selectivity for H
2
S removal, and removal of the sulfur compounds COS, CS
2
, and
mercaptans. As a rule, these sulfur compounds exist at much lower concentrations
than do H
2
S, but their removal is important if product specifications have an upper
limit on total sulfur. Note that in cases of deep liquids recovery, most of the COS,
CS
2
, and mercaptans present in the feed will stay with the liquid product (NGL).The
Table 5.2, adapted from Echterhoff (1991), presents a detailed summary of
and Schaaf (1977) and Figure 5.1 to Figure 5.4 are based on their recommen-
© 2006 by Taylor and Francis Group, LLC
96 Fundamentals of Natural Gas Processing
FIGURE 5.1 Process selection chart for CO
2
removal with no H
2
S present. (Adapted from
Tennyson and Schaaf, 1977.)
FIGURE 5.2 Process selection chart for H
2
S removal with no CO
2
present. (Adapted from
Tennyson and Schaaf, 1977.)
10
1
10
2
10
0
10
−1
10
0
10
1
10
2
10
3
Partial pressure of acid gas in product, psia
P
a
r
t
i
a
l

p
r
e
s
s
u
r
e

o
f

a
c
i
d

g
a
s

i
n

f
e
e
d
,

p
s
i
a Physical solvent
L
i
n
e

o
f

e
q
u
a
l

i
n
l
e
t

a
n
d

o
u
t
l
e
t

p
r
e
s
s
u
r
e
s
Physical solvent or activated hot potassium carbonate
Physical solvent,
hybrid, or hot
potassium
carbonate
Hybrid
Amine
Activated hot potassium carbonate
or inhibited concentrated amine
Activated hot potassium carbonate or amine
10
0
10
1
10
−1
10
−2
10
0
10
1
10
2
10
3
Partial pressure of acid gas in product, psia
P
a
r
t
i
a
l

p
r
e
s
s
u
r
e

o
f

a
c
i
d

g
a
s

i
n

f
e
e
d
,

p
s
i
a
L
i
n
e

o
f

e
q
u
a
l

i
n
l
e
t

a
n
d

o
u
t
l
e
t
p
r
e
s
s
u
r
e
s
Physical solvent or hybrid
High loading DEA,
hybrid or physical
solvent
Physical solvent
Amine, stretford or hybrid
ADIP, hybrid or high loading DEA
Stretford
© 2006 by Taylor and Francis Group, LLC
Gas Treating 97
FIGURE 5.3 Process selection chart for simultaneous H
2
S and CO
2
removal. (Adapted
from Tennyson and Schaaf, 1977.)
FIGURE 5.4 Process selection chart for selective H
2
S removal with CO
2
present. (Adapted
from Tennyson and Schaaf, 1977.)
10
1
10
2
10
0
10
−1
10
0
10
1
10
2
10
3
Partial pressure of acid gas in product, psia
P
a
r
t
i
a
l

p
r
e
s
s
u
r
e

o
f

a
c
i
d

g
a
s

i
n

f
e
e
d
,

p
s
i
a
High loading DEA.
or physical
solvent
Physical solvent
Physical solvent
Activated hot potassium carbonate,
hybrid or amine
Hybrid
or amine
L
i
n
e

o
f

e
q
u
a
l

i
n
l
e
t

a
n
d

o
u
t
l
e
t

p
r
e
s
s
u
r
e
s
10
1
10
2
10
0
10
−1
10
0
10
1
10
2
10
3
Partial pressure of acid gas in product, psia
P
a
r
t
i
a
l

p
r
e
s
s
u
r
e

o
f

a
c
i
d

g
a
s

i
n

f
e
e
d
,

p
s
i
a
Physical solvent
L
i
n
e

o
f

e
q
u
a
l

i
n
l
e
t

a
n
d

o
u
t
l
e
t

p
r
e
s
s
u
r
e
s
Tertiary amine, sterically hindered amine or mixed amine
Stretford, tertiary amine
or mixed amine
© 2006 by Taylor and Francis Group, LLC
98 Fundamentals of Natural Gas Processing
TABLE 5.2
CO
2
and H
2
S Removal Processes for Gas Streams
Process
Normally
Capable of
Meeting H
2
S
Specification
a

Removes
COS,
CS
2
, and
Mercaptans
Selective
H
2
S
Removal
Minimum
CO
2
Level
Obtainable
Solution
Subject to
Degradation?
(Degrading Species)
Monoethanol-
amine (MEA)
Yes Partial No 100 ppmv at
low to
moderate
pressures
Yes (COS,
CO
2
, CS
2
, SO
2
, SO
3

and mercaptans)
Diethanol amine
(DEA)
Yes Partial No 50 ppmv
in SNEA-
DEA
process
Some (COS, CO
2
CS
2
,
HCN and
mercaptans)
Triethanol amine
(TEA)
No Slight No Minimum
partial
pressure
of 0.5 psia
(3 kPa)
Slight (COS, CS
2
and
mercaptans)
Methyldiethanol-
amine
(MDEA)
Yes Slight Some Bulk
removal
only
No
Diglycol amine
(DGA)
Yes Partial No 100 ppmv at
moderate
to high
pressures
Yes (COS, CO
2
,
and CS
2
)
Diisopropanol-
amine (DIPA)
Yes COS only Yes Not
applicable
Resistant to
degradation by COS
Sulfinol Yes Partial Yes
(Sulfinol
-M)
50 ppmv,
50%
slippage
while
meeting
H
2
S
product
spec
Some (CO
2
and CS
2
)
Hot potassium
carbonate
Yes, with
special design
features
Partial No Not
reported
Not reported
Stretford Yes No Yes No
significant
amounts of
CO
2
are
removed
Yes (CO
2
at high
concentrations)
© 2006 by Taylor and Francis Group, LLC
Gas Treating 99
are susceptible to degradation by O
2
, which forms undesirable and corrosive
compounds.
In this discussion, we consider only the more commonly used amine processes,
an alkali salt process (hot potassium carbonate), the Selexol

physical absorption
process, molecular sieve adsorption, the Ryan/Holmes cryogenic fractionation pro-
cess, membranes, and H
2
S scavengers. For a more comprehensive discussion of
acid gas purification, the reader should refer to Kohl and Nielsen (1997) and
Engineering Data Book (2004b).
5.2 SOLVENT ABSORPTION PROCESSES
In solvent absorption, the two major cost factors are the solvent circulation rate,
which affects both equipment size and operating costs, and the energy require-
and disadvantages of chemical and physical solvents.
TABLE 5.2 (Continued)
CO
2
and H
2
S Removal Processes for Gas Streams
Process
Normally
Capable of
Meeting H
2
S
Specification
a

Removes
COS,
CS
2
, and
Mercaptans
Selective
H
2
S
Removal
Minimum
CO
2
Level
Obtainable
Solution
Subject to
Degradation?
(Degrading Species)
Selexol

Yes Slight Some Can be
slipped or
absorbed
No
Rectisol Yes Yes No 1 ppmv Not reported
Molecular sieves Yes Yes
(excluding
CS
2
)
Some Can meet
cryogenic
spec when
CO
2
feed
content is
less than
~2%
Not applicable
Membranes No Slight No Feed
concentra-
tion
dependent
Not applicable
a
H
2
S specification is 25% grain H
2
S per 100 scf (6 mg/m
3
)
Source: Adapted from Echterhoff, 1991.
ment for regenerating the solvent. Table 5.3 summarizes some of the advantages
distribution of these sulfur compounds is discussed in Chapter 10. The amines
© 2006 by Taylor and Francis Group, LLC
100 Fundamentals of Natural Gas Processing
5.2.1 AMINES
Amines are compounds formed from ammonia (NH
3
) by replacing one or more
of the hydrogen atoms with another hydrocarbon group. Replacement of a single
hydrogen produces a primary amine, replacement of two hydrogen atoms pro-
duces a secondary amine, and replacement of all three of the hydrogen atoms
produces a tertiary amine. Primary amines are the most reactive, followed by
the secondary and tertiary amines. Sterically hindered amines are compounds
in which the reactive center (the nitrogen) is partially shielded by neighboring
groups so that larger molecules cannot easily approach and react with the
nitrogen. The amines are used in water solutions in concentrations ranging from
approximately 10 to 65 wt% amines, and a table of their physical properties is
are amines with OH groups attached to the hydrocarbon groups to reduce their
processing.
Amines remove H
2
S and CO
2
in a two step process:
1. The gas dissolves in the liquid (physical absorption).
2. The dissolved gas, which is a weak acid, reacts with the weakly
basic amines.
Absorption from the gas phase is governed by the partial pressure of the H
2
S
and CO
2
in the gas, whereas the reactions in the liquid phase are controlled by
TABLE 5.3
Comparison of Chemical and Physical Solvents
Chemical Solvents
Advantages Disadvantages
Relatively insensitive to H
2
S and CO
2

partial pressure
High energy requirements for regeneration
of solvent
Can reduce H
2
S and CO
2
to ppm levels Generally not selective between
CO
2
and H
2
S
Amines are in a water solution, and thus the
treated gas leaves saturated with water
Physical Solvents
Advantages Disadvantages
Low energy requirements for
regeneration
May be difficult to meet H
2
S specifications
Can be selective between H
2
S and CO
2
Very sensitive to acid gas partial pressure
volatility. Figure 5.5 shows the formulas for the common amines used in gas
included in Appendix B. All commonly used amines are alkanolamines, which
© 2006 by Taylor and Francis Group, LLC
Gas Treating 101
the reactivity of the dissolved species. The principal reactions (see, e.g., Sigmund
et al., 1981) are summarized in the next section.
5.2.1.1 Basic Amine Chemistry
Figure 5.5 shows the structure of the commonly used amines.
Amines are bases, and the important reaction in gas processing is the ability of
the amine to form salts with the weak acids formed by H
2
S and CO
2
in an aqueous
solution. When a gas stream that contains the H
2
S, CO
2
, or both, is contacted by a
primary or secondary amine solution, the acid gases react to form a soluble acid−
base complex, a salt, in the treating solution. The reaction between the amine and
both H
2
S and CO
2
is highly exothermic. Regardless of the structure of the amine,
H
2
S reacts rapidly with the primary, secondary, or tertiary amine via a direct proton-
transfer reaction, as shown in Equation 5.1, to form the amine hydrosulfide:
R
1
R
2
R
3
N + H
2
S ↔ R
1
R
2
R
3
NH
+
HS

(5.1)
The reaction is shown for a tertiary amine but applies to primary and secondary
amines as well. The reaction between the amine and the CO
2
is more complex
because CO
2
reacts via two different mechanisms. When dissolved in water, CO
2
hydrolyzes to form carbonic acid, which, in turn, slowly dissociates to bicarbonate.
FIGURE 5.5 Molecular structures of commonly used amines.
Primary Amines
N – CH
2
– CH
2
– OH
H
H
Monoethanolamine (MEA)
N – (CH
2
)
2
– O – (CH
2
)
2
– OH
H
Diglycolamine (DGA)
H
Secondary Amines
HO – (CH
2
)
2
– N – (CH
2
)
2
– OH
H
Diethanolamine (DEA)
HO – CH –CH
2
– N – CH
2
– CH – OH
Diisopropanolamine (DIPA)
CH
2
H CH
2
Tertiary Amines
HO – (CH
2
)
2
– N – (CH
2
)
2
– OH
HO – (CH
2
)
2
Triethanolamine (TEA)
HO – (CH
2
)
2
– N – (CH
2
)
2
– OH
Methyldiethyanolamine (MDEA)
CH
3
Sterically Hindered Amine
N – C – CH
2
– OH
CH
3
H
2-amino-2methyl-1-propanol (AMP)
H
CH
3
© 2006 by Taylor and Francis Group, LLC
102 Fundamentals of Natural Gas Processing
The bicarbonate then undertakes an acid−base reaction with the amine to yield
the overall reaction shown by Equation 5.2:
CO
2
+ H
2
Ο ↔ H
2
CO
3
(carbonic acid) (5.2a)
H
2
CO
3
↔ H
+
+ HCO
3

(bicarbonate) (5.2b)
H
+
+ R
1
R
2
R
3
N ↔ R
1
R
2
R
3
NH
+
(5.2c)
CO
2
+ H
2
O + R
1
R
2
R
3
N ↔ R
1
R
2
R
3
NH
+
HCO
3

(5.2)
This acid−base reaction occurs with any of the alkanolamines, regardless of
the amine structure, but the reaction is not as rapid as that of H
2
S, because the
carbonic acid dissociation step to the bicarbonate is relatively slow.
A second CO
2
reaction mechanism, shown by Equation 5.3, requires the
presence of a labile (reactive) hydrogen in the molecular structure of the amine.
CO
2
+ R
1
R
2
NH ↔ R
1
R
2
N
+
HCOO

(5.3a)
R
1
R
2
N
+
HCOO

+ R
1
R
2
NH ↔ R
1
R
2
NCOO

+ R
1
R
2
NH
2
+
(5.3b)
CO
2
+ 2R
1
R
2
NH ↔ R
1
R
2
NH
2
+
R
1
R
2
NCOO

(5.3)
This second reaction mechanism for CO
2,
which forms the amine salt of a
substituted carbamic acid, is called the carbamate formation reaction and occurs
only with the primary and secondary amines. The CO
2
reacts with one primary
or secondary amine molecule to form the carbamate intermediate, which in turn
reacts with a second amine molecule to form the amine salt. The rate of CO
2
reaction via carbamate formation is much faster than the CO
2
hydrolysis reaction,
but slower than the H
2
S acid−base reaction. The stoichiometry of the carbamate
reaction indicates that the capacity of the amine solution for CO
2
is limited to
0.5 moles of CO
2
per mole of amine if the only reaction product is the amine
carbamate. However, the carbamate undergoes partial hydrolysis to form bicar-
bonate, which regenerates free amine. Hence, CO
2
loadings greater than 0.5, as
seen in some plants that employ diethanolamine (DEA), are possible through the
hydrolysis of the carbamate intermediate to bicarbonate. We found no systematic
study to identify how to obtain the higher loadings.
The fact that CO
2
absorption occurs by two reaction mechanisms with different
kinetic characteristics, significantly affects the relative absorption rates of H
2
S and
CO
2
among the different alkanolamines. For primary and secondary amines, little
difference exists between the H
2
S and CO
2
reaction rates because of the availability
of the rapid carbamate formation for CO
2
absorption. Therefore, the primary and
secondary amines achieve essentially complete removal of H
2
S and CO
2
. However,
because the tertiary amines have no labile hydrogen, they cannot form the
© 2006 by Taylor and Francis Group, LLC
Gas Treating 103
carbamate. Tertiary amines must react with CO
2
via the slow hydrolysis mechanism
in Equation 5.2. With only the slow acid−base reaction available for CO
2
absorption,
methyldiethanolamine (MDEA) and several of the formulated MDEA products
yield significant selectivity toward H
2
S relative to CO
2
, and, consequently, all of
the H
2
S is removed while some of the CO
2
“slips” through with the gas. Because
the CO
2
reaction with water to form bicarbonate is slow and the H
2
S reaction is
fast for MDEA, the H
2
S reaction is considered gas-phase limited and the CO
2
reaction is considered liquid-phase limited.
For the reactions discussed above, high pressures and low temperatures drive
the reactions to the right, whereas high temperatures and low pressures favor the
reverse reaction, which thus provides a mechanism for regeneration of the amine
solution
Accurate modeling of the vapor−liquid phase behavior of CO
2
and H
2
S in amine
solutions for process design requires incorporation of the ionic chemistry into the
above equations. As the demand for lower sulfur contents increased, modeling
became more important. The Gas Processors Association (GPA) is conducting a
multiyear research program to provide the data needed to improve amine modeling,
including the widely used model proposed by Kent and Eisenberg (1976). For
details of the work, the reader should review the list of GPA research reports
available through the GPA or in the Engineering Data Book (2004b).
Brief descriptions of the more commonly used amines follow; more detailed
discussions are available in the handbook of Kohl and Nielson (1997) and the
Engineering Data Book (2004b). Table 5.4 provides an abbreviated list of the
more important operating parameters for the commonly used amines.
5.2.1.2 Monoethanolamine
Monoethanolamine (MEA) is the most basic of the amines used in acid treating
and thus the most reactive for acid gas removal. It has the advantage of a high
solution capacity at moderate concentrations, and it is generally used for gas
TABLE 5.4
Some Representative Operating Parameters for Amine Systems
MEA DEA DGA MDEA
Wt% amine 15 to 25 25 to 35 50 to 70 40 to 50
Rich amine acid gas loading
Mole acid gas/mole amine
0.45 to 0.52 0.43 to 0.73 0.35 to 0.40 0.4 to 0.55
Acid gas pickup
Mole acid gas/mole amine
0.33 to 0.40 0.35 to 0.65 0.25 to 0.3 0.2 to 0.55
Lean solution residual acid gas
Mole acid gas/mole amine
0.12 ± 0.08 ± 0.10 ± 0.005 to 0.01
Source: Engineering Data Book (2004b).
© 2006 by Taylor and Francis Group, LLC
104 Fundamentals of Natural Gas Processing
streams with moderate levels of CO
2
and H
2
S when complete removal of both
impurities is required.
Monoethanolamine has a number of disadvantages, including:
• A relatively high vapor pressure that results in high vaporization losses
• The formation of irreversible reaction products with COS and CS
2

• A high heat of reaction with the acid gases that results in high energy
• The inability to selectively remove H
2
S in the presence of CO
2

• Higher corrosion rates than most other amines if the MEA concen-
tration exceeds 20% at high levels of acid gas loading (Kohl and
Nielsen, 1997)
• The formation of corrosive thiosulfates when reacted with oxygen
(McCartney, 2005)
A slow production of “heat stable salts” form in all alkanol amine solutions,
primarily from reaction with CO
2
. Oxygen enhances the formation of the salts.
In addition to fouling regenerator reboilers, high concentrations of salts can carry
over to the contactor and cause foaming, which degrades contactor efficiency.
One “advantage” of MEA (and DGA) over other amines is that “reclaimers” are
installed in-line for intermittently removing these salts, along with the irreversible
reaction products formed from COS and CS
2
Reclaiming).
5.2.1.3 Diglycolamine
Compared with MEA, low vapor pressure allows Diglycolamine [ 2-(2-aminoethoxy)
ethanol] (DGA) to be used in relatively high concentrations (50 to 70%), which
results in lower circulation rates. It is reclaimed onsite to remove heat stable salts
and reaction products with COS and CS
2
.
5.2.1.4 Diethanolamine
Diethanolamine (DEA), a secondary amine, is less basic and reactive than MEA.
Compared with MEA, it has a lower vapor pressure and thus, lower evaporation
losses; it can operate at higher acid gas loadings, typically 0.35 to 0.8 mole acid
gas/mole of amine versus 0.3 to 0.4 mole acid-gas/mole; and it also has a lower
energy requirement for reactivation. Concentration ranges for DEA are 30 to 50 wt%
and are primarily limited by corrosion. DEA forms regenerable compounds with
COS and CS
2
and, thus, can be used for their partial removal without significant
solution loss. DEA has the disadvantage of undergoing irreversible side reactions
with CO
2
and forming corrosive degradation products; thus, it may not be the best
choice for high CO
2
gases. Removal of these degradation products along with the
heat stable salts must be done by use of either vacuum distillation or ion exchange.
The reclaiming may be done offsite or in portable equipment brought onsite.
requirements for regeneration (see Table 5.5)
(see Section 5.2.1.10 Amine
© 2006 by Taylor and Francis Group, LLC
Gas Treating 105
5.2.1.5 Methyldiethanolamine
Methyldiethanolamine (MDEA), a tertiary amine, selectively removes H
2
S to pipe-
line specifications while “slipping” some of the CO
2
. As noted previously, the CO
2
slippage occurs because H
2
S hydrolysis is much faster than that for CO
2
,

and the
carbamate formation reaction does not occur with a tertiary amine. Consequently,
short contact times in the absorber are used to obtain the selectivity. MDEA has a
low vapor pressure and thus, can be used at concentrations up to 60 wt% without
appreciable vaporization losses. Even with its relatively slow kinetics with CO
2
,
MDEA is used for bulk removal of CO
2
from high-concentration gases because
energy requirements for regeneration are lower than those for the other amines. It
is not reclaimable by conventional methods (Veroba and Stewart, 2003).
5.2.1.6 Sterically Hindered Amines
In acid gas removal, steric hindrance involves alteration of the reactivity of a
primary or secondary amine by a change in the alkanol structure of the amine.
A large hydrocarbon group attached to the nitrogen shields the nitrogen atom and
hinders the carbamate reaction. The H
2
S reaction is not significantly affected by
amine structure, because the proton is small and can reach the nitrogen. However,
CO
2
removal can be significantly affected if the amine structure hinders the fast
carbamate formation reaction and allows only the much slower bicarbonate for-
mation (reactions 5.2a and 5.2b).
5.2.1.7 Mixed Amines
The selectivity of MDEA can be reduced by addition of MEA, DEA, or propri-
etary additives. Thus, it can be tailored to meet the desired amount of CO
2
slippage
and still have lower energy requirements than do primary and secondary amines.
The following section looks at the difference in heats of reaction.
5.2.1.8 Heats of Reaction
The magnitude of the exothermic heats of reaction, which includes the heat of
solution, of the amines with the acid gases is important because the heat liberated
in the reaction must be added back in the regeneration step. Thus, a low heat of
marizes the important data for the common amines. The values in Table 5.5 are
approximate because heats of reaction vary with acid gas loading and solution
concentration. Jou et al. (1994) and Carson et al. (2000) report measured heats
of reaction for CO
2
in MEA, DEA, and MDEA.
5.2.1.9 Process Flow Diagram
A typical diagram for the removal of acid gases from natural gas by use of MEA
representative, not definitive. The sour gas feed enters the bottom of the contactor
reaction translates into smaller energy regeneration requirements. Table 5.5 sum-
is shown in Figure 5.6. The diagram is simplified, and all operating conditions are
© 2006 by Taylor and Francis Group, LLC
106 Fundamentals of Natural Gas Processing
at pressures to 1,000 psi (70 bar) and temperatures in the range of 90°F (32°C).
The sour gas flows upward, countercurrent to the lean amine solution which flows
down from the top. The lean amine that returns to the contactor is maintained at a
temperature above the vapor that exits the contactor to prevent any condensation
TABLE 5.5
Average Heats of Reaction
a
of the Acid Gases
in Amine Solutions
Amine H
2
S, Btu/lb (kJ/kg) CO
2
, Btu/lb (kJ/kg)
MEA 610 (1420) 825 (1920)
DEA 555 (1290) 730 (1700)
DGA
®
674 (1570) 850 (1980)
MDEA 530 (1230) 610 (1420)
a
The heats of reaction include both heat of solution and heat
of reaction.
Source: Engineering Data Book (2004b).
FIGURE 5.6 Process flow diagram for amine treating by use of MEA. Contactor com-
monly operates at pressures up to 1,000 psi (70 bar). Flow rates to reclaimer are 1 to 3%
of amine circulation rate. (Adapted from Smith, 1996.)
Contactor
Regenerator
Sweet gas
Sour gas
Lean
amine
Rich
amine
Air cooled
heat exchanger
Flash tank
Gas
Liquid
F
i
l
t
e
r
Lean-rich
heat exchanger
Acid gas
Reboiler
Reclaimer
Steam
Steam
Waste
Reflux
accumulator
140°F
75 to 100
psig
100°F
280 to 300°F
180 to
200°F
Air cooled
heat exchanger
90°F
220°F
12 psig
8 psig
225 to 260°F
© 2006 by Taylor and Francis Group, LLC
Gas Treating 107
of heavier liquid hydrocarbons. Intimate contact between the gas and amine solution
is achieved by use of either trays or packing in the contactor.
The contactor operates above ambient temperature because of the combined
exothermic heat of absorption and reaction. The maximum temperature is in
the lower portion of the tower because the majority of the absorption and
reaction occurs near the bottom of the unit. The temperature “bulge” in the
tower can be up to about 180°F (80°C). The treated gas leaves the top of the tower
water saturated and at a temperature controlled by the temperature of the lean amine
that enters, usually around 100°F (38°C).
The rich amine leaves the bottom of the unit at temperatures near 140°F
(60°C) and enters a flash tank, where its pressure is reduced to 75 to 100 psig (5
to 7 barg) to remove (flash) any dissolved hydrocarbons. The dissolved hydro-
carbons are generally used as plant fuel. If necessary, a small stream of lean
amine is contacted with the fuel gas to reduce the H
2
S concentration. The rich
amine then passes through a heat exchanger and enters the solvent regenerator
(stripper) at temperatures in the range of 180 to 220°F (80 to 105°C). The reboiler
on the stripper generally uses low-pressure steam. The vapor generated at the
bottom flows upward through either trays or packing, where it contacts the rich
amine and strips the acid gases from the liquid that flows down. A stream of lean
amine is removed from the stripper, cooled to about 110°F (45°C), and reenters
the contactor at the top to cool and condense the upward-flowing vapor stream.
The vapor, which consists mostly of acid gases and water vapor, exits the top of
the stripper and is generally processed for sulfur recovery.
The lean amine exits the bottom of the stripper at about 260°F (130°C) and
is pumped to the contactor pressure, exchanges heat with the rich amine stream,
and is further cooled before it enters the top of the contactor. Both the treated
gas that leaves the gas contactor and the acid gas to the Claus unit are water
saturated.
equipment and the operating conditions are also representative of DEA and
DGA units, with the exception of the reclaimer. DEA units do not have reclaim-
ers, and DGA reclaimers operate at slightly higher temperatures of 375 to 385°F
(190 to 195°C).
Lyddon and Nguyen (1999) evaluate several alterations to the basic flow
scheme for amine units that can increase gas throughput. When additional
capacity exists in the regenerator but not in the absorber, and additional pressure
drop is acceptable, static mixers may help. Blending sour gas and lean amine
provides essentially one additional absorber stage. Two configurations were
considered. In one case, a side stream of sour gas passes through a static mixer.
Following separation from the amine, the gas is blended with the sweet gas
from the absorber. In the other case, the whole stream goes through the static
mixer upstream of the inlet separator before it enters the absorber. In both cases,
the rich amine from the static mixer goes to the regenerator. Lyddon and Nguyen
(1999) report one case in which use of the bypass static mixers allowed a 17%
increase in plant capacity.
Figure 5.6 shows an MEA unit. Smith (1996) states the major pieces of
© 2006 by Taylor and Francis Group, LLC
108 Fundamentals of Natural Gas Processing
5.2.1.10 Amine Reclaiming
Amines react with CO
2
and contaminants, including oxygen, to form organic acids.
These acids then react with the basic amine to form heat stable salts (HSS). As their
name implies, these salts are heat stable, accumulate in the amine solution, and must
be removed. For MEA and DGA solutions, the salts are removed through the use
process, as described by Smith (1996). The reclaimer is filled with lean amine, and
a strong base, such as sodium carbonate or sodium hydroxide, is added to the solution
to neutralize the heat stable salts. A slipstream of 1 to 3% of the circulating amine
is then continuously added to the reclaimer while the mixture is heated. Water and
amine vapor are taken off the top, which leaves the contaminants in the liquid
bottoms. Heating is continued until the temperature is approximately 300°F (150°C)
for MEA or 360 to 380°F (180 to 195°C) for DGA. The cycle is then stopped and
the bottoms that contain the contaminants (dissolved salts, suspended solids) are
removed. DEA does not form a significant amount of nonregenerable degradation
products, and it requires more difficult reclaiming through vacuum distillation or ion
exchange. Further details on reclaimers can be found in the Engineering Data Book
(2004b), Smith (1996), Simmons (1991), and Holub and Sheilan (2000).
5.2.1.11 Operating Issues
A comprehensive discussion of operating issues is given by Holub and Sheilan
(2000) and is beyond the scope of this book. The brief discussion here of their
work highlights a few of the most important considerations.
Corrosion—Some of the major factors that affect corrosion are:
• Amine concentration (higher concentrations favor corrosion)
• Rich amine acid gas loading (higher gas loadings in the amine favor
corrosion)
• Oxygen concentration
• Heat stable salts (higher concentrations promote corrosion and foaming)
In addition to destroying vessels and piping, the corrosion products can cause
foaming.
Solution Foaming—Foaming of the liquid amine solution is a major problem
because it results in poor vapor−liquid contact, poor solution distribution, and
solution holdup with resulting carryover and off spec gas. Among the causes of
foaming are suspended solids, liquid hydrocarbons, surface active agents, such
as those contained in inhibitors and compressor oils, and amine degradation
products, including heat stable salts. One obvious cure is to remove the offending
materials; the other is to add antifoaming agents.
Heat Stable Salts—As mentioned above, these amine degradation products
can cause both corrosion and foaming. They are normally dealt with through the
of a reclaimer (refer to Figure 5.6), which utilizes a semicontinuous distillation
use of amine reclaimers (see Section 5.2.1.10, Amine Reclaiming).
© 2006 by Taylor and Francis Group, LLC
Gas Treating 109
5.2.2 ALKALI SALTS
The hot potassium carbonate process for removing CO
2
and H
2
S was developed
by the United States Bureau of Mines and is described by Benson and coworkers
in two papers (Benson et al., 1954, 1956). Although the process was developed
for the removal of CO
2
, it can also remove H
2
S if H
2
S is present with CO
2
. Special
designs are required for removing H
2
S to pipeline specifications or to reduce CO
2
to low levels.
The process is very similar in concept to the amine process, in that after
physical absorption into the liquid, the CO
2
and H
2
S react chemically with the
solution. The chemistry is relatively complex, but the overall reactions are rep-
resented by
K
2
CO
3
+ CO
2
+ H
2
O ↔ 2KHCO
3
(5.4a)
K
2
CO
3
+ H
2
S ↔ KHS + KHCO
3
(5.4b)
The process has several variations, and a simplified diagram of the single-
stage process is shown in Figure 5.7.
In a typical application, the contactor will operate at approximately 300 psig
(20 barg), with the lean carbonate solution entering near 225°F (110°C) and leaving
FIGURE 5.7 Process flow diagram for hot potassium carbonate process. (Adapted from
Engineering Data Book, 2004b.)
Sour
feed gas
Sweet
product gas
Rich
solution
Lean
solution
Acid gas
Absorber
Steam
Condenser
Stripper
© 2006 by Taylor and Francis Group, LLC
110 Fundamentals of Natural Gas Processing
at 240°F (115°C). The rich carbonate pressure is reduced to approximately 5 psig
(0.3 barg) as it enters the stripper. Approximately one third to two thirds of the
absorbed CO
2
is released by the pressure reduction, reducing the amount of steam
required for stripping (Kohl and Nielsen, 1997). The lean carbonate solution leaves
the stripper at the same temperature as it enters the contactor, and eliminates the
need for heat exchange between the rich and lean streams. The heat of solution for
absorption of CO
2
in potassium carbonate is small, approximately 32 Btu/cu ft of
CO
2
(Benson et al., 1954), and consequently the temperature rise in the contactor is
small and less energy is required for regeneration. Kohl and Nielson (1997) give a
thorough discussion of the process and its many modifications.
5.3 PHYSICAL ABSORPTION
In the amine and alkali salt processes, the acid gases are removed in two steps:
physical absorption followed by chemical reaction. In processes such as Selexol

or Rectisol

, no chemical reaction occurs and acid gas removal depends entirely
on physical absorption. Some of the inherent advantages and disadvantages of
physical absorption processes are summarized below:
• Absorption processes are generally most efficient when the partial
pressures of the acid gases are relatively high, because partial pressure
is the driving force for the absorption.
• Heavy hydrocarbons are strongly absorbed by the solvents used, and
consequently acid gas removal is most efficient in natural gases with
low concentrations of heavier hydrocarbons.
• Solvents can be chosen for selective removal of sulfur compounds,
which allows CO
2
to be slipped into the residue gas stream and reduce
separation costs.
• Energy requirements for regeneration of the solvent are lower than in
systems that involve chemical reactions.
• Separation can be carried out at near-ambient temperature.
• Partial dehydration occurs along with acid gas removal, whereas amine
processes produce a water saturated product stream that must be dried
in most applications.
Selexol

is a typical application of physical absorption and a number of open
literature articles describe the process. Consequently, it was selected as an example
to describe the absorption process. The handbook of Kohl and Nielsen (1997)
provides information on other absorption processes.
5.3.1 SOLVENT PROPERTIES
Selexol

is a polyethylene glycol and has the general formula:
CH
3
—O—CH
2
—(CH
2
—O—CH
2
)
N
—CH
2
—O—CH
3
© 2006 by Taylor and Francis Group, LLC
Gas Treating 111
Because Selexol

is a mixture of homologues (the value of N in the general
formula for Selexol

varies), the physical properties discussed below are approximate.
Table 5.6 (Sweny, 1980) presents R
K
, the ratio of the K-value* for methane,
K
CH4
(arbitrarily assigned a value of 1), to the K-values of the other component,
R
K
= K
CH4
/K
component

K
greater than unity indicates that the solubility of the component in Selexol

is greater
than that of methane, whereas a value less than unity indicates the opposite. The
values should be regarded as only representative because pressure and temperature
are not specified and, as previously noted, the composition of Selexol

is variable.
Because R
K
for CO
2
and H
2
S are 15 and 134, respectively, these gases are
preferentially absorbed (relative to CH
4
), and, consequently, physical absorption is
an effective technique for acid gas removal. The process can reduce H
2
S to 4 ppmv,
reduce CO
2
to levels below 50 ppmv, and essentially remove all mercaptans, CS
2
,
and COS. Two additional features of Table 5.6 are worth mentioning. Because the
R
K
values for hydrocarbons heavier than CH
4
are fairly high (6.4 for C
2
H
6
, 15.3 for
C
3
H
8
, and 35 for n-C
4
H
10
), Selexol

will remove substantial quantities of these
hydrocarbons, a feature that can be either positive or negative, depending on the
composition of the gas being processed and the desired products. Finally, the R
K
value
of H
2
O is extremely high and consequently, Selexol

provides some dehydration.
Bucklin and Schendel (1985) reported gas solubility data in the form of
absorption coefficients (volume of gas absorbed per volume of liquid). We con-
verted values for some of the compounds into scf gas absorbed/gal Selexol

and
TABLE 5.6
Typical Relative Ratio of K-values
Component R
K
Component R
K
CH
4
1 H
2
S 134
C
2
H
6
6.4 C
6
H
14
165
CO
2
15 CH
3
SH 340
C
3
H
8
15.3 C
7
H
16
360
Isobutane 28 CS
2
360
n-Butane 35 SO
2
1400
COS 35 C
6
H
6
3800
Isopentane 67 C
4
H
4
S 8100
n-Pentane 83 H
2
O 11000
R
K
= (K
CH4
/K
Component
) in Selexol

solvent for various
solutes, where K
CH4
= 1.
Source: Sweny (1980).
* The K-value is the ratio of the mole fraction of the component in the vapor phase (y) to its mole
fraction in the liquid phase (x), K = y/x. High K-values indicate the material is predominately in the
vapor phase, whereas low K-values indicate a higher concentration in the liquid phase (x).
. Table 5.7 lists some physical properties of Selexol . An R value
© 2006 by Taylor and Francis Group, LLC
112 Fundamentals of Natural Gas Processing
provides approximate solubility at higher pressures. The lines also ignore prob-
able interaction between solutes. Epps (1994) discusses the limited solubility of
hydrocarbons in the liquid (when the liquid is saturated, a second liquid hydrocarbon
phase forms above the Selexol

phase) and the effect of absorbed water on the
solubility of hydrocarbons.
5.3.2 REPRESENTATIVE PROCESS CONDITIONS
Applications of Selexol

are varied and, consequently, no common process flow
diagrams are available. Sweny (1980) presents flow diagrams for nine different
applications, and Epps (1994) discusses plants for the dehydration of natural gas
and hydrocarbon dew point control.
One plant discussed by Epps (1994), designated only as a European distri-
because it is a modern application, and both inlet and outlet gas compositions
2
TABLE 5.7
Representative Property Data for Selexol

Molar mass (approximate) = 280
Flash point (Cleveland open cup) = 304°F (151°C)
Freezing point
a
= −8 to −20°F (−22 to −28°C)
Vapor pressure at 25°C = <0.01 mm Hg
Specific heat at 77°F (25°C) = 0.49 Btu/lbm °F
= 2.0 kJ/kg K
Density at 77°F (25°C) = 1030 kg/m3 (8.60 lb/gal)
Viscosity at 77°F (25°C) = 5.8 cp (5.8 × 10
−3
Ns/m
2
)
Thermal conductivity at 77°F (25°C) = 0.11 Btu/(h)(ft)(°F)
= 0.19 W/m K
Surface tension at 77°F (25°C) = 34.3 dynes/cm
Heat of solution at 77°F, Btu/lb of solute (kJ/kg)
CO
2
= 160 (372)
H
2
S = 190 (442)
CH
4
= 75 (174)
Odor = very mild
Toxicity = nil
a
Slush appears at −8°F complete solidification at −20°F
Sources: Sweny and Valentine (1970) and Clare and Valentine (1975).
* For an ideal system, Henry’s law assumes a linear relation between the solubility of gas component
i and its partial pressure, y
i
P = k
i
x
i
where k
i
is the Henry’s constant.
were reported (Table 5.8). The plant pretreats the gas to reduce CO , ethane,
plotted them in Figure 5.8. The figure assumes a Henry’s law relationship,* which
bution plant, is shown in Figure 5.9. This plant was selected for discussion
© 2006 by Taylor and Francis Group, LLC
Gas Treating 113
FIGURE 5.8 Solubility of various gases in Selexol

solvent at 70°F (21°C) as a function
of partial pressure. (Based on data by Sweny and Valentine, 1970.)
FIGURE 5.9 Process schematic for a Selexol

gas treating facility. (Adapted from Epps,
1994.)
0
10
20
30
200 400 600
Gas partial pressure, psia
S
o
l
u
b
i
l
i
t
y
,

s
c
f
/
g
a
l

s
e
l
e
x
o
l
CH
4
CO
2
H
2
S
COS
C
H
3
S
H
High
pressure
flash
Low
pressure
flash
Compressor
Absorber
Refrigeration
Knockout
drum
Chilled
inlet gas
Treated
gas Reject gas
(mostly hydrocarbons)
Vacuum
flash
Vacuum
unit
Solvent
© 2006 by Taylor and Francis Group, LLC
114 Fundamentals of Natural Gas Processing
and heavier hydrocarbon levels before final purification in molecular sieve units
and subsequent liquefaction. The plant is designed to process 26 MMscfd (0.74
MMSm
3
/d) entering the Selexol

unit at 603 psia (41.6 bar) and 32°F (0°C). The
lean solvent, cooled to 25°F (−3.9°C) with propane

refrigerant, enters the absorber
where it absorbs CO
2
and some of the ethane and heavier hydrocarbons. The rich
solvent from the absorber is regenerated by reduction of the pressure in three flash
drums, from 603 to 106 psia (41.6 to 7.3 bar) in the high-pressure drum, from 106
to 16 psia (7.3 to 1.1 bar) in the medium-pressure drum, and from 16 to 3 psia (1.1
to 0.21 bar) in the vacuum drum. Lean Selexol

from the vacuum drum is recom-
pressed and sent to the propane

chiller. The treated gas that leaves the absorber
passes through a knockout drum and filter separator to remove entrained Selexol

and condensed hydrocarbons. Table 5.8 shows that the treated gas meets the spec-
ifications of a maximum of 0.50% CO
2
and a maximum of 6.5% ethane

and heavier
hydrocarbons. In addition, the water content of the gas is reduced from 75 ppmv
to 12 ppmv, H
2
S is reduced from 2 ppmv to essentially nothing, and methyl
mercaptan is reduced from 5 ppmv to 1 ppmv. Unlike the amine systems, no
irreversible products are generated in the process, which thus eliminates the need
for reclaiming.
Other detailed descriptions of acid gas removal with Selexol

are given by
Hegwar and Harris (1970), Johnson (1984), Judd (1978), Mortko (1984), Oberding
et al. (2004), Raney (1976), and Shah (1989).
5.3.3 HYBRID PROCESSES
systems. To take advantage of the strengths of each type, a number of hybrid
processes commercially used, and under development, combine physical solvents
TABLE 5.8
Composition of Inlet and Outlet Gas in a Selexol

Unit
Gas Composition (Mol%)
Component Inlet Outlet
Hydrogen sulfide 0.0002 —
Methyl mercaptan 0.00050 0.00012
Carbon dioxide 2.44 0.29
Nitrogen 0.785 0.88
Methane 86.317 93.02
Ethane 7.539 5.33
Propane 2.403 0.35
Heavier
hydrocarbons
0.515 0.130
Source: Epps (1994).
Table 5.3 showed the strengths and weaknesses of amine and physical solvent
© 2006 by Taylor and Francis Group, LLC
Gas Treating 115
with amines (Engineering Data Book, 2004b). Depending upon the solvent−amine
combination, nearly complete removal of H
2
S, CO
2
, and COS is possible. Other
hybrid systems provide high H
2
S and COS removal while slipping CO
2
. Sulfinol
®
currently is one of the more commonly used processes. The process uses a
combination of a physical solvent (sulfolane) with DIPA or MDEA. The selected
amine depends upon the acid gases in the feed and whether CO
2
removal is
required. Like the physical solvent processes, the hybrid systems may absorb
more hydrocarbons, including BTEX, but that property can be adjusted by varying
water content.
5.4 ADSORPTION
Acid gases, as well as water, can be effectively removed by physical adsorption
on synthetic zeolites. Applications are limited because water displaces acid gases
in dehydration.) Figure 5.10, which shows typical isotherms for CO
2
and H
2
S on
molecular sieve, indicates that at ambient temperatures substantial quantities of
both gases are adsorbed even at low partial pressures.
Molecular sieve can reduce H
2
S levels to the 0.25 gr/100 scf (6 mg/m
3
)
specification. However, this reduction requires regeneration of the bed at 600°F
(315°C) for extended time (Engineering Data Book, 2004b) with the potential
for COS formation if 4A* is used.
FIGURE 5.10 Adsorption of H
2
S (
__
) and CO
2
(- - ) on type 5A molecular sieve at 77°F
(25°C). (adapted from Linde, undated).
* Molecular sieves are classified by their nominal pore diameter in Angstroms. See Chapter 6 for
details.
0
5
10
15
20
25
0.00 0.25 0.50 0.75 1.00 1.25 1.50
Gas pressure, bar
L
b

a
d
s
o
r
b
e
d
/
1
0
0

l
b

a
d
s
o
r
b
a
n
t
on the adsorbent bed. (Chapter 6 provides more details on adsorption and its use
© 2006 by Taylor and Francis Group, LLC
116 Fundamentals of Natural Gas Processing
Chi and Lee (1973) studied the coadsorption of H
2
S, CO
2
, and H
2
O on a
5A molecular sieve from a natural gas mixture under a variety of conditions.
Figure 5.11, from their paper, shows a typical concentration versus time curve.
In the figure, y is the concentration in the exit stream and y
o
is the concentration
in the inlet to the bed. The gas that entered the bed was saturated with H
2
O
and contained both CO
2
(1.14 mol%) and H
2
S (0.073 mol%.). Because the CO
2
content of the gas was 15.6 times that of the H
2
S, the bed quickly saturated
with CO
2
, and its breakthrough was almost instantaneous. As the H
2
S was
adsorbed and moved down the column, it displaced the CO
2
and, consequently,
after approximately 30 minutes, the CO
2
exit concentration peaked at a value
greater than its inlet concentration.
The same phenomenon occurs when the H
2
S is displaced by the water.
Because the H
2
S must be removed to extremely low levels, the bed is effectively
exhausted from the perspective of H
2
S purification shortly after H
2
S breakthrough
occurs and, thus, well before the bed is totally saturated. Chi and Lee (1973)
present empirical correlations for prediction of dynamic saturation capacity and
breakthrough times for H
2
S under a variety of operating conditions.
Detailed information on adsorber design calculations is available in the paper
by Cummings and Chi (1977) and the handbook by Kohl and Nielsen (1997).
The Engineering Data Book (2004b) notes that a key point in design is to properly
design for treatment of the regeneration gas because the peak H
2
S concentration
may be 30 times the H
2
S concentration in the feed.
2
gas. The configuration is similar to that for dehydration but with the significant
difference that the regeneration gas contains high quantities of H
2
S as well as
FIGURE 5.11 Effluent H
2
S and CO
2
concentration from adsorption bed as a function of
time (adapted from Chi and Lee, 1973). See text for details.
Time, hours
0.5
1.5
4 8
0
21 22
y
/
y
0
1.0
CO
2
H
2
S
H
2
O
Figure 5.12 shows a typical flow diagram for removal of H S from natural
water as it leaves the adsorbent bed and, thus, must be treated. (Chapter 6 presents
© 2006 by Taylor and Francis Group, LLC
Gas Treating 117
a thorough discussion of adsorption.) The flow configuration shows the first bed
in the adsorption cycle, the second bed cooling down after regeneration, and the
third bed undergoing regeneration with hot gas.
5.5 CRYOGENIC FRACTIONATION
Distillation is the most widely used process to separate liquid mixtures, and at
first glance it seems a good prospect for removing CO
2
and H
2
S from natural
gas, because the vapor pressures of the principal components are considerably
CO
2
from methane, CO
2
from ethane, and CO
2
from H
2
S. These difficulties are
summarized below (Engineering Data Book, 2004 b).
CO
2
from methane: Relative volatilities (K
C1
/K
CO2
) at typical distillation
conditions are about 5 to 1. Therefore one would expect simple fractionation
to work. However, because the liquid CO
2
phase freezes when it becomes
concentrated, the practical maximum-vapor concentration of methane is only
85 to 90 mol%.
CO
2
from ethane: In addition to solidification problems, CO
2
and ethane form
an azeotrope (liquid and vapor compositions are equal) and consequently, a
shows an isotherm at −64°F (−53°C) for the system, and the azeotrope appears
at approximately 0.6 mole fraction of CO
2
.
FIGURE 5.12 Schematic of integrated natural gas desulfurization plant. Dashed line
denotes regeneration gas stream. (Adapted from Engineering Data Book, 2004b.)
1 2 3
Inlet
gas
Liquid
Molecular sieve
beds
Sweet gas
Acid gas removal
H
2
S
Regeneration gas recycle
different (Figure 5.13). However, problems are associated with the separation of
complete separation of these two by simple distillation is impossible. Figure 5.14
© 2006 by Taylor and Francis Group, LLC
118 Fundamentals of Natural Gas Processing
2 2
mixture forms a pinch at high CO
2
concentrations. This separation by conventional
distillation is complicated by the need to have an overhead product that has roughly
100 ppmv H
2
S if the stream is vented. The bottoms product should contain less
than two-thirds CO
2
,

assuming the stream is feed to a Claus unit.
A number of techniques are available for solving these problems, but the
Ryan/Holmes process (Holmes et al., 1982) is probably the most widely used.
FIGURE 5.13 Vapor pressure of light natural gas components (NIST, 2005).
FIGURE 5.14 Vapor−liquid equilibrium curve for CO
2
and C
2
H
6
at −64°F (−53°C).
Azeotropic composition is 0.60 mol fraction CO
2
. (Data from Brown et al., 1988.)
0
20
40
60
80
100
50 100 150 200 250 300 350 400
Temperature, K
V
a
p
o
r

p
r
e
s
s
u
r
e
,

b
a
r
CH
4
CO
2
H
2
S
C
2
H
6
0.0
2.0
4.0
6.0
8.0
10.0
0.0 0.2 0.4 0.6 0.8 1.0
Mol fraction CO
2
P
r
e
s
s
u
r
e
,

b
a
r
CO from H S: This distillation is difficult because, as Figure 5.15 shows, the
© 2006 by Taylor and Francis Group, LLC
Gas Treating 119
This process is an extractive distillation process* that uses hydrocarbons to
significantly alter the behavior of the system and thus, effectively eliminate the
distillation problems. The hydrocarbons are normally mixtures of propane and
heavier hydrocarbons obtained from the feed mixture. As a result, no additional
separations are required. Readers are referred to the original papers for details.
5.6 MEMBRANES
Membranes are used in natural gas processing for dehydration, fuel-gas condi-
tioning, and bulk CO
2
removal, but presently CO
2
removal is by far the most
important application. We first present the fundamentals of membrane separations
and then discuss their application to CO
2
removal.
5.6.1 MEMBRANE FUNDAMENTALS
From thermodynamics, the driving force for movement through the membrane
is the difference in chemical potential, m, for a given component on the two sides
of the membrane. If subscript i is the diffusing component, then
m
i,feed
> m
i,permeate

FIGURE 5.15 Vapor−liquid equilibrium curve for CO
2
and H
2
S at 20 atm. (Data from
Bierlein and Kay, 1953.)
* Extractive distillation makes distillation of close boiling components possible by addition of a
solvent to the mixture to alter the relative volatility of the two key components. The products from
the distillation include one of the keys at high purity and a mix of the other key plus the solvent.
This mixture is fractionated in another column for recovery of the solvent and production of the pure
second key.
0.0
0.2
0.4
0.6
0.8
1.0
0.0 0.2 0.4 0.6 0.8 1.0
CO
2
in liquid phase, mol fraction
C
O
2


i
n

v
a
p
o
r

p
h
a
s
e
,

m
o
l

f
r
a
c
t
i
o
n
© 2006 by Taylor and Francis Group, LLC
120 Fundamentals of Natural Gas Processing
where the material is moving from feed side to the permeate side. To make the
equation physically comprehensible in gas systems, fugacity, f, which is propor-
tional to the chemical potential is used and gives
f
i,feed
> f
i,permeate
(5.5)
If ideal behavior is assumed for the diffusing gas, then the fugacities can be
replaced by the partial pressures and
y
i,feed
P
i,feed
> y
i,permeate
P
i,permeate
, (5.6)
where y is the mole fraction and P is the total pressure. This equation shows what
terms affect the driving force across the membrane. The equation can be rear-
ranged to obtain
. (5.7)
This relation tells us that the separation achieved can never
exceed the pressure ratio .
Because the discussion above was based on classical thermodynamics, it tells
us nothing about the rate at which the diffusion processes take place. A discussion
of diffusion rate requires use of Fick’s law. Fick’s law for solution-diffusion
membranes in rectangular coordinates is (e.g., Echt et al., 2002)
J
i
= (S
i
D
i
p
i
)/L, (5.8)
where J is the flux of component i, that is, the molar flow of component i through
the membrane per unit area of membrane, S
i
is the solubility term, D
i
is the
diffusion coefficient, is the partial pressure difference across the membrane,
and L is the thickness of the membrane. Customarily, S
i
, and D
i
are combined
into a single term, the permeability, P
i
, and thus divides Fick’s law into two parts,
P
i
/L, which is membrane dependent and , which is process dependent. Note
that P
i
/L is not only dependent on the membrane but also dependent on operating
conditions, because S
i
and D
i
depend on both temperature and pressure. P
i
also
depends weakly upon the composition of the gases present.
All the mixture components have a finite permeability, and the separation is
based upon differences in them. Customarily, selectivity,
1-2
, is used, which is
the ratio of two permeabilities, P
1
/P
2
,

a term important in process design and
evaluation. An of 20 for CO
2
/CH
4
means that CO
2
moves through the membrane
20 times faster than does methane.
y
y
P
P
i permeate
i feed
feed
permeate
,
,

















( / )
, ,
y y
i permeate i feed
( / ) P P
feed permeate
∆p
i
∆p
i
α
α
© 2006 by Taylor and Francis Group, LLC
Gas Treating 121
The above discussion introduces the basic principles regarding membrane
separation. Readers desiring more depth should consult the engineering handbook
of Kohl and Nielsen (1997).
5.6.2 CARBON DIOXIDE REMOVAL FROM NATURAL GAS
Many different types of membranes have been developed or are under devel-
opment for industrial separations (Echt et al., 2002; Baker, 2002), but for
CO
2
removal, the industry standard is presently cellulose acetate. These
membranes are of the solution-diffusion type, in which a thin layer (0.1 to
0.5 µm) of cellulose acetate is on top of a thicker layer of a porous support
material. Permeable compounds dissolve into the membrane, diffuse across
it, and then travel through the inactive support material. The membranes are
thin to maximize mass transfer and, thus, minimize surface area and cost, so
the support layer is necessary to provide the needed mechanical strength.
Commercial membrane configurations are either hollow fiber elements or flat
sheets wrapped into spirally wound elements. Presently, about 80% of gas-
separation membranes are formed into hollow fiber modules (Baker, 2002), like
those shown in Figure 5.16. The low-pressure, bore-feed configuration is a
FIGURE 5.16 Cutaway view of the two module configurations used with hollow fiber
membranes. (Courtesy of Membrane Research and Technology, Inc.)
High pressure, shell-side gas feed module
Low pressure, bore-side gas feed module
Hollow fibers
Resin
layer
O-ring
Vessel
Feed
Residue
Residue
End plate
Permeate
Permeate
Feed
End plate
Hollow fiber layer
O-ring
© 2006 by Taylor and Francis Group, LLC
122 Fundamentals of Natural Gas Processing
countercurrent flow configuration similar to a shell-tube heat exchanger with the
gas entering on the tube side. It has the advantage of being more resistant to
fouling because the inlet gas flows through the inside of the hollow fibers.
However, the mechanical strength of the membrane limits the pressure drop across
the membrane. The configuration is only used in low-pressure applications, such
as air separation and air dehydration (Baker, 2002).
To handle high pressures, the permeate flows into the hollow fiber from the
shell side. This feature makes the membrane much more susceptible to plugging,
and gas pretreatment is usually required (Baker, 2002). The gas flow is cross
current and provides good feed distribution in the module. This configuration is
widely used to remove CO
2
from natural gas.
In the spiral wound element shown in Figure 5.17, two membrane sheets are
separated by a permeate spacer and glued shut at three ends to form an envelope or
leaf. Many of these leaves, separated by feed spacers, are wrapped around the
permeate tube, with the open end of the leaves facing the tube. Feed gas travels
along the feed spacers, the permeating species diffuse through the membranes
and down the permeate spacers into the permeate tube, and the residue gas exits
at the end. The gas flow is cross flow in this configuration.
The spiral configuration is inherently more resistant than the hollow fiber
membranes to trace components that would alter the polymer permeability. It also
allows a wider range of membrane materials to be used. However, the hollow fiber
membranes are cheaper to fabricate, and thus dominate the field (Baker, 2002).
Once the elements have been manufactured, they are grouped into modules, as
cylindrical tubes contains the membranes, and the separate unit on the right of the
picture is the feed-pretreatment unit, described in more detail in Section 5.6.3.5,
Feed Gas Pretreatment. Echt et al. (2002) and Kohl and Nielsen (1997) provide
more complete details of the construction of elements and modules.
FIGURE 5.17 Cutaway view of spiral wound membrane element. (Copyright 2002 UOP
LLC. All rights reserved. Used with permission.)
Feed
High CO
2
NG
Permeate
CO
2
-rich
gas
Feed spacer
Membrane
Permeate spacer
Membrane
Feed spacer
Low CO
2
product
Low CO
2
product
shown in Figure 5.18. These modules are then mounted on a skid to make a complete
unit, as shown in Figure 5.19. The unit on the left of the picture composed of long
© 2006 by Taylor and Francis Group, LLC
Gas Treating 123
5.6.3 OPERATING CONSIDERATIONS
5.6.3.1 Flow Patterns
and 5.21 show simplified examples for the CH
4
/CO
2
separation. In the single-stage
unit, the overall methane recovery is only 90.2%, but the process requires flow
FIGURE 5.18 Cutaway view of spiral wound membrane module. (Copyright 2002 UOP
LLC. All rights reserved. Used with permission.)
FIGURE 5.19 Carbon dioxide membrane skid installed in gas plant.
Membrane
housing
Permeate tube
union
Permeate
Feed
Inlet seat
(U-cup)
Membrane
element
Residual
A number of different flow patterns can be used with membranes. Figures 5.20
© 2006 by Taylor and Francis Group, LLC
124 Fundamentals of Natural Gas Processing
through only one membrane, and no recompression is needed. To increase methane
recovery to 98.7%, a two-stage unit requires recompression of the first-stage per-
meate. Greater levels of methane recovery are obviously possible by application of
three or more stages, but additional elements can quickly become uneconomical
because of both membrane cost and recompression energy required.
FIGURE 5.20 A non-optimized single-stage CO
2
/CH
4
membrane separation process
(Spillman and Cooley, 1989).
FIGURE 5.21 A non-optimized two stage CO
2
/CH
4
membrane separation process (Spillman
and Cooley, 1989).
Feed (A)
Residue or retenate (B)
Membrane
Permeate (C)
Composition (mole %)
Methane
Carbon dioxide
Flow rate (MMscfd)
Pressure (psig)
A C
93.0
7.0
20.00
850
98.0
2.0
17.11
835
63.4
36.6
2.89
10
Methane recovery = 90.2%
Single-stage process (non-optimized)
B
Two-stage process (non-optimized)
Feed (A)
Residue (B)
(retenate)
Membrane
Membrane
(D)
(C)
(E)
Permeate
Compressor
Composition (mole %)
Methane
Carbon dioxide
Flow rate (MMscfd)
Pressure (psig)
A B C D
93.0
7.0
20.00
850
98.0
2.0
18.74
835
18.9
81.1
1.26
10
63.4
36.6
3.16
10
93.0
7.0
1.90
850
Methane recovery = 98.7%
E
© 2006 by Taylor and Francis Group, LLC
Gas Treating 125
The second law of thermodynamics dictates that energy is required for a
separation. For membrane processes, this law translates into loss of pressure.
However, in many processes, the cost of recompression, if needed, still makes
membranes an attractive separation process.
5.6.3.2 Flow Rate
A maximum acceptable feed gas rate per unit area applies to the membrane, and
required membrane area is directly proportional to the flow rate. Membrane units
perform well at reduced feed rates, but their performance drops when design flow
rates are exceeded. Additional modules are added in parallel to accept higher
flow rates.
5.6.3.3 Operating Temperature
Increased operating temperature increases permeability but decreases selectivity.
Because membranes are organic polymers, they have a maximum operating
temperature that depends upon the polymer used. Exceeding this temperature will
degrade membrane material and shorten the useful life of the unit.
5.6.3.4 Operating Pressures
Increased feed pressure decreases both the permeability and selectivity, but at the
same time, the pressure difference across the membrane (the driving force in the
membrane flux equation) is increased, which results in a net increase in flow through
the membrane. The same effect of increasing the driving force is achieved by
lowering the permeate pressure. Echt et al. (2002) point out that the pressure ratio
across the membrane is an important design parameter, and thus, design engineers
try to maintain the lowest possible permeate pressure (see Equation 5.7).
5.6.3.5 Feed Gas Pretreatment
Because membranes are susceptible to degradation from impurities, pretreatment
is usually required. The impurities possibly present in natural gas that may cause
damage to the membrane (Echt et al., 2002) include:
• Liquids. The liquids may be entrained in the feed to the unit or formed
by condensation within the unit. Liquids can cause the membrane to
swell, which results in decreased flux rates and possible membrane
damage. Liquids can form internally by two mechanisms: (1) because
of condensation of higher molar mass compounds caused by the cool-
ing that occurs (Joule-Thomson effect) as the gas expands to a lower
pressure through the membrane, and (2) because CO
2
and the lighter
hydrocarbons diffuse more quickly than the heavier hydrocarbons, the
dew point of the nondiffusing gas may increase to the point where
condensation occurs.
© 2006 by Taylor and Francis Group, LLC
126 Fundamentals of Natural Gas Processing
• High-molar-mass hydrocarbons (C
15
+) such as compressor lube oils.
These compounds coat the membrane surface and result in a loss of
performance. The concentrations are low but the effect is cumulative.
• Particulates. These materials block the small flow passages in the
membrane element. Erosion of the membrane could also be a problem.
• Corrosion inhibitors and well additives. Certain of these compounds
are destructive to membrane material.
A common method for pretreating the feed gas to a membrane system is
shown in Figure 5.22. The coalescing filter removes any entrained liquids; the
adsorbent bed takes out trace contaminants such as volatile organic compounds
(VOC); the particulate filter removes any dust from the adsorbent bed; and the
heater superheats the gas to prevent liquid formation in the membrane unit. The
system shown has the following disadvantages (Echt et al., 2002):
• The adsorbent bed is the only unit that removes heavy hydrocarbons.
Consequently, if the gas contains more heavy hydrocarbons than anti-
cipated, or in the event of a surge of these materials, the adsorbent bed
may become saturated in a relatively short time, and thus allow heavy
hydrocarbons to contact the membrane.
• Only the heater provides superheat, and, consequently, if this unit fails, the
entire membrane system must be shut down. Echt et al (2002) discuss other
pretreatment methods that address the disadvantages discussed above.
5.6.4 ADVANTAGES AND DISADVANTAGES
OF MEMBRANE SYSTEMS
The pros and cons of commercial membrane systems are discussed by Kohl and
Nielsen (1997) and Echt et al. (2002). The following summarizes their conclusions:
Advantages
• Low capital investment when compared with solvent systems
• Ease of operation: process can run unattended
FIGURE 5.22 Schematic of membrane-pretreating equipment. (Adapted from Echt et
al., 2002.)
Coalescing
filter
Adsorbent
guard bed
Particle filter
Feed
To
membrane
Heater
© 2006 by Taylor and Francis Group, LLC
Gas Treating 127
• Ease of installation: Units are normally skid mounted
• Simplicity: No moving parts for single-stage units
• High turndown: The modular nature of the system means very high
turndown ratios can be achieved
• High reliability and on-stream time
• No chemicals needed
• Good weight and space efficiency
Disadvantages
• Economy of scale: Because of their modular nature, they offer little
economy of scale
• Clean feed: Pretreatment of the feed to the membrane to remove
particulates and liquids is generally required
• Gas compression: Because pressure difference is the driving force
for membrane separation, considerable recompression may be
required for either or both the residue and permeate streams
For natural gas systems, the following disadvantages also apply:
• Generally higher hydrocarbon losses than solvent systems
• H
2
S removal: H
2
S and CO
2
permeation rates are roughly the same, so
H
2
S specifications may be difficult to meet
• Bulk removal: Best for bulk removal of acid gases; membranes alone
cannot be used to meet ppmv specifications
For offshore applications, membranes are attractive because of unattended
operation, no chemicals needed or generated, and good weight and space efficiency.
5.7 NONREGENERABLE HYDROGEN
SULFIDE SCAVENGERS
When the quantity of sulfur to be recovered is small, on the order of 400 lb/d
(180 kg/d) or less, the processes discussed previously are uneconomical, and
small-scale batch processes are used for H
2
S removal. These processes generally
use a nonregenerable scavenger. Many such technologies are in general use, and
According to Houghton and Bucklin (1994), the most widely used of the
scavengers are those that utilize iron oxides, especially iron sponge. In addition
to the references mentioned above, detailed discussions of all the above processes
may be found in Veroba and Stewart (2003), Engineering Data Book (2004b),
and Kohl and Nielsen (1997). Detailed economic comparisons of the most popular
scavenger processes may be found in Houghton and Bucklin (1994). The Gas
Technology Institute markets a PC program (GRI Scavenger Calcbase™, Gas
Research Institute) that makes approximate cost comparisons for a variety of
scavengers.
a summary is shown in Table 5.9.
© 2006 by Taylor and Francis Group, LLC
128 Fundamentals of Natural Gas Processing
Although the iron sponge is a commonly used solid, nonregenerative scavenger,
liquid oxidation−reduction (redox) processes use iron as an oxidizing agent in
solution (Engineering Data Book, 2004b). The process involves four basic steps:
1. Removal of H
2
S from the gas by absorption into a caustic solution
2. Oxidation of the HS

ion to elemental sulfur via the oxidizing agent
3. Separation and removal of sulfur from the solution
4. Regeneration (i.e., oxidation) of the oxidizing agent by use of air
While the oxidizing agent forms the desired reaction, it also reacts with the
sulfur species to form metal sulfides that precipitate from solution. To avoid the
metal sulfide reaction, the solution contains chelating agents (organic compounds
that bind with the metal ion to restrict its reactivity but still permit electron transfer
for oxidation−reduction reactions). A common chelating agent for iron is EDTA
(ethylenediaminetetraacetic acid). Several processes, including Lo-Cat II
®
, SulF-
erox
®
, and Sulfint-HP, use iron and chelating agents.
5.8 BIOLOGICAL PROCESSES
A biological process for removing H
2
S from natural gas has been reported by
Cline et al. (2003). In this process, the gas stream that contains the H
2
S is first
absorbed into a mild alkaline solution, and the absorbed sulfide is oxidized to
elemental sulfur by naturally occurring microorganisms.
Cline et al. (2003) reported the successful startup and operation of a commercial
plant in Canada. The plant is designed to treat a high-pressure natural gas and
produce a product that meets H
2
S specifications (4 ppmv or lower). The sulfur plant
capacity is approximately 1 ton per day, with a sulfur removal efficiency of 99.5%
or higher. Cline et al. (2003) offer the following claims regarding the process:
• Cost effective up to 50 tons/d (50 tonnes/d) of sulfur
• H
2
S concentrations in the sour gas from 50 ppmv to 100 vol%
TABLE 5.9
H
2
S Scavenger Processes
Solid-based processes: Iron oxides
Zinc oxides
Liquid-based processes: Nitrites
Amine-aldehyde condensates
Caustic scrubbing
Aldehydes
Oxidizers
Metal-oxide slurries
Source: Adapted from Holub and Sheilan (2000).
© 2006 by Taylor and Francis Group, LLC
Gas Treating 129
• Sour gas pressures to 75 barg (1,100 psig)
• H
2
S concentrations in the sweet gas 1 ppmv
• Formation of hydrophilic sulfur that prevents equipment fouling or blocking
5.9 SAFETY AND ENVIRONMENTAL
CONSIDERATIONS
5.9.1 AMINES
The obvious safety concern with amine treating is the potential for H
2
S leaks in
the plant, even from spilled rich amine. In addition, some sections operate at high
temperatures. Caustic handling is another hazard if MEA reclaiming is performed
on-site. Reclaimer waste products are toxic and must be handled with care
(Engineering Data Book, 2004b).
From an environmental perspective, in addition to the remote chance of
hydrogen sulfide release, amines have an affinity for BTEX (benzene, toluene,
ethylbenzene, and xylenes), which may be vented during amine regeneration
are used with reclaimers, the reclaimer solids present a disposal problem,
especially with MEA because caustic or soda ash is added to help reverse the
reactions.
5.9.2 PHYSICAL ABSORPTION
When H
2
S or other sulfur compounds are removed from a gas stream that contains
high levels of these materials, the potential always exists for a leak. Depending
upon the process, the solvent may be hazardous or toxic.
5.9.3 ADSORPTION
Safety and environmental problems associated with adsorbents such as molecular
sieve are relatively minor. Thorough regeneration and purging must be done
before the adsorbent can be replaced. However, it should be nonhazardous and
disposable in a land fill.
Most solid scavengers are respiratory and eye irritants. Spent iron sponge is
pyrophoric, and great care must be taken in the removal and disposal of reacted
iron-sponge material (Veroba and Stewart, 2003). The manufacturer’s recommen-
dations for this material must be carefully followed to prevent a serious incident.
5.9.4 MEMBRANES
Membranes are probably the safest and most environmentally friendly of the
processes for gas treating. No chemicals are used, no waste disposal by-products
are generated, and membranes operate at low pressures and generally ambient
temperatures.
if the sulfur is not recovered. Chapter 9 provides more details. If MEA or DGA
© 2006 by Taylor and Francis Group, LLC
130 Fundamentals of Natural Gas Processing
REFERENCES
Baker, R.W., Future directions of membrane gas separation technology, Ind. Eng. Chem.
Res. 41, 1391, 2002.
Baker, R.W., Wijmans, J.G., and Kaschemekat, J.H., The design of vapor-gas separation
systems, J. Membrane Sci. 151, 55, 1998.
Benson, H.E., Field, J.H., and Jimeson, R.M., CO
2
absorption employing hot potassium
carbonate solutions, Chem. Eng. Progress, 50, 356, 1954.
Benson, H.E., Field, J.H., and Haynes, W.P., Improved process for CO
2
absorption uses
hot carbonate solutions, Chem. Eng. Progress, 52, 433, 1956.
Bierlein, J.A. and Kay, W.B., Phase equilibrium properties of system carbon dioxide-
hydrogen sulfide, Ind. Eng. Chem., 45, 618, 1953.
Brown, T.S., Kidnay, A.J., and Sloan, E.D, Vapor-liquid equilibria in the carbon dioxide
+ ethane system, Fluid Phase Equil., 40, 169, 1988.
Bucklin, R.W. and Schendel, R.L., Comparison of physical solvents used for gas process-
ing, in Acid and Sour Gas Treating Processes, Newman, S.A., Ed., Gulf Publishing
Company, Houston, TX, 1985. (As quoted in Kohl, A. and Nielsen, R., Gas
Purification, 5th ed., Gulf Publishing, Houston, TX, 1997, 1197.
Carson, J.K., Marsh, K.N., and Mather, A.E., Enthalpy of solution of carbon dioxide in
(water + monoethanolamine, or diethanolamine, or N-methyldiethanolamine) and
(water + monoethanolamine + N-methyldiethanolamine) at T = 298.15K, J. Chem.
Thermo., 32, 1285, 2000.
Chi, C. W. and Lee, H., Natural gas purification by 5A molecular sieves and its design
method, AIChE Symp. Ser., 69, 95, 1973.
Clare, R.T. and Valentine, J.P., Acid Gas Removal Using the Selexol

Process, Proceeding
of the Second Quarterly Meeting of the Canadian Natural Gas Processors Asso-
ciation, Edmonton, Alberta, Canada, June 5, 1975.
Cline, C., Hoksberg, A., Abrey, R., and Janssen, A., Biological Process for Removal
from Gas Streams, The Shell-Paques/THIOPAQ Gas Desulfurization Process,
Proceedings of the Laurance Reid Gas Conditioning Conference, Norman, OK,
2003, 1.
Cummings, W. and Chi, C.W., Natural Gas Purification with Molecular Sieves, Canadian
Chemical Engineering Conference, Calgary, Alberta, Canada, October 23−26,
1977.
Echt, W.I., Dortmund, D.D., and Malino, H.M., Fundamentals of Membrane Technology
for CO
2
Removal from Natural Gas, Proceedings of the Laurance Reid Gas
Conditioning Conference, Norman, OK, 2002, 1.
Echterhoff, L.W., State of the Art of Natural Gas Processing Technologies, Gas Research
Institute Report GRI - 91/0094, 1991.
Engineering Data Book, 12th ed., Sec. 2, Product Specifications, Gas Processors Supply
Association, Tulsa, OK, 2004a.
Engineering Data Book, 12th ed., Sec. 21, Hydrocarbon Treating, Gas Processors Supply
Association, Tulsa, OK, 2004b.
Epps, R, Use of Selexol

solvent for Hydrocarbon Dewpoint Control and Dehydration of
Natural Gas, Proceedings of the Laurance Reid Gas Conditioning Conference,
Norman, OK, 1994, 26.
Hegwar, A.M. and Harris, R.A, Selexol

solves high H
2
S/CO
2
problem, Hydrocarbon
Process., 49 (4) 103, 1970.
Gas Research Institute, www.gastechnology.org, Retreived July 2005.
© 2006 by Taylor and Francis Group, LLC
Gas Treating 131
Holmes, A.S., Ryan, J.M., Price, B.C., and Styring, R.E., Pilot Tests Prove Ryan/Holmes
Cryogenic Acid Gas/Hydrocarbon Separations, Proceedings of the Sixty-First
Annual Convention of the Gas Processors Association, Tulsa, OK, 1982, 75.
Holub, P.E. and Sheilan, M., Fundamentals of Gas Sweetening, Proceedings of the
Laurance Reid Gas Conditioning Conference, Norman, OK, 2000.
Houghton, J.E. and Bucklin, R.W., Nonregenerable H
2
S Scavenger Update, Proceedings
of the Laurance Reid Gas Conditioning Conference, Norman, OK, 1994, 110.
Johnson, J.E., Selexol

solvent process reduces lean, high-CO
2
natural gas treating costs,
Energy Progr., 4 (4) 241, 1984.
Jou, F.-Y, Otto, F.D., and. Mather, A.E, Vapor liquid equilibrium of carbon dioxide in
aqueous mixtures of monoethanolamine and methyldiethanolamine, Ind. Eng.
Chem. Res. 33, 2002, 1994.
Judd, D.K., Selexol

unit saves energy, Hydrocarbon Process., 57 (4) 122, 1978.
Kent, R.L. and Eisenberg, B., Better data for amine treating, Hydrocarbon Process., 55
(2) 87, 1976.
Klinkenbijl, J.M., Dillon, M.L., and Heyman, E.C., Gas Pre-Treatment and Their Impact
on Liquefaction Processes, Proceedings of the Seventy-Eighth Annual Convention
of the Gas Processors Association, Tulsa, OK, 1999, 299.
Kohl, A. and Nielsen, R., Gas Purification, 5th ed., Gulf Publishing, Houston, TX, 1997.
Kopperson, D., Horne, S., Kohn, G., Romansky, D., Chan, C., and Duckworth, G.L., Acid
gas disposal 1: Injecting acid gas with water creates new disposal option, Oil Gas
J., 96 (31) 33, 1998.
Kopperson, D., Horne, S., Kohn, G., Romansky, D., Chan, C., and Duckworth, G.L., Two
gases illustrate acid gas/water injection scheme, Oil Gas J., 96 (32) 64, 1998.
Linde Molecular Sieves Non-Hydrocarbon Materials Data Sheets, Form 9691-F.
Lyddon, L. and Nguyen, H., Analysis of Various Flow Schemes for Sweetening with
Amines, Proceedings of the Seventy-Eighth Annual Convention of the Gas Pro-
cessors Association, Tulsa, OK, 1999, 177.
McCartney, D.G., private communication, 2005.
Meyer, H.S. Volume and Distribution of Subquality Natural Gas in the United States,
GasTIPS, Gas Research Institute, Winter 2000, p.10.
Moritis, G., New companies, infrastructure, projects reshape landscape for CO
2
EOR in
US, Oil Gas J., 99 (20) 68, 2001.
Mortko, R.A., Remove H
2
S selectively, Hydrocarbon Process., 63 (6) 78, 1984.
NIST, National Institute of Standards and Technology, U.S. Department of Commerce,
Thermophysical properties of fluid systems, 2005, webbook.nist.gov/chemistry/
fluid/, Retrieved August 2005.
Oberding, W., Goff, R., Townsend, M., Chapin, D., Naulty, D., and Worah, V., Lessons
Learned and Technology Improvements at the Lost Cabin Gas Plant, Proceedings
of the Laurance Reid Gas Conditioning Conference, Norman, OK, 2004, 273.
Raney, D.R., Remove carbon dioxide with Selexol

, Hydrocarbon Process., 55 (4) 73, 1976.
Shah, V.A., Integrated Gas Treating and Hydrocarbon Recovery Process Using Selexol

Solvent Technology, Proceedings of the Sixty-Eighth Annual Convention of the
Gas Processors Association, Tulsa, OK, 1989, 197.
Simmons, C.V., Reclaiming Used Amine and Glycol Solutions, Proceedings of the Laurance
Reid Gas Conditioning Conference, Norman, OK, 1991, 335.
Sigmund, P.W., Butwell, K.F., and Wussler, A.J., The H
2
S Process, an Advanced Process
for Selective H
2
S Removal, Proceedings of the Sixtieth Annual Convention of the
Gas Processors Association, Tulsa, OK, 1981.
© 2006 by Taylor and Francis Group, LLC
132 Fundamentals of Natural Gas Processing
Smith, W.A., Good Operating Practices for Amine Treating Systems, Proceedings of the
Laurance Reid Gas Conditioning Conference, Norman, OK, 1996, 303.
Spillman, R.W. and Cooley, T.E., Membrane Gas Treating, Proceedings of the Eighty-
Eighth Annual Convention of the Gas Processors Association, Tulsa, OK, Tulsa,
Oklahoma, 1989, 186.
Sweny, J.W., High CO
2
-H
2
S Removal with Selexol

Solvent, Proceedings of the Sixty-
Ninth Annual Convention of the Gas Processors Association, Tulsa, OK, 1980,163.
Sweny, J.W. and Valentine, J.P., Physical solvent stars in gas treatment/purification, Chem-
ical Eng., 54 (19) 77 1970.
Tennyson, R.N. and Schaaf, R.P., Guidelines can help choose proper process for gas-
treating plants, Oil Gas J., 75 (2) 78, 1977.
Veroba, R. and Stewart, E., Fundamentals of Gas Sweetening, Proceedings of the Laurance
Reid Gas Conditioning Conference, Norman, OK. 2003, 1.
Wichert, E. and Royan, T., Acid gas injection eliminates sulfur recovery expense, Oil Gas
J., 95 (17) 1997.
© 2006 by Taylor and Francis Group, LLC
133
6
Gas Dehydration
6.1 INTRODUCTION
Typically, dehydration is important in three areas:
• Gas gathering. Water needs to be removed to reduce pipeline corrosion
and eliminate line blockage caused by hydrate formation. The water dew
point should be below the lowest pipeline temperature to prevent free
Field Liquids Removal
Gas & Liquids from Wells
Inlet
Compression
Sulfur
Recovery
Gas Treating
Dehydration
Nitrogen
Rejection
Helium
Recovery
Outlet
Compression
Elemental
Sulfur
Field Dehydration
Field Acid Gas Removal
Field Compression
Sales Gas
Water &
Solids
Natural
Gas Liquids
Natural
Gasoline
Raw Helium
CO
2
Liquefaction
Liquefied
Natural Gas
Inlet Receiving
Hydrocarbon
Recovery
Liquids Processing
water formation. Chapter 3 discusses field operations and gas hydrates.
© 2006 by Taylor and Francis Group, LLC
134 Fundamentals of Natural Gas Processing
• Product dehydration. Both gas and liquid products have specifications
on water content. Sales gas that leaves a plant is usually dry if cryogenic
hydrocarbon liquid recovery is used. Liquid and gas streams may be
water saturated after amine treatment or coming from underground
storage. Most product specifications, except for propane, require that
no free water be present (Engineering Data Book, 2004a). This require-
ment puts the maximum water in sales gas at 4 to 7 lb/MMscf (60 to
110 mg/Sm
3
). For liquids, the water content is 10 to 20 ppmw. Com-
• Hydrocarbon recovery. Most plants use cryogenic processes to recover
the C
2
+ fraction from inlet gas. If acid gases are removed by use of
amine processes, the exit gas leaves water saturated. To prevent hydrate
formation in the cryogenic section of hydrocarbon recovery, the water
carbon recovery processes.
The first two areas are less demanding and a variety of processes are available
to meet the need. However, the third application, as well as gas liquefaction and
cryogenic nitrogen rejection units, requires water contents an order of magnitude
lower than the other two. Molecular sieves can provide dehydration to this level.
Water content is stated in a number of ways:
• Mass of water per volume of gas, lb/MMscf (mg/Sm
3
)
• Dew point temperature, °F (°C), which is the point that liquid water, real
or hypothetically subcooled, will condense out of the hydrocarbon phase
• Concentration, parts per million by volume (ppmv)
• Concentration, parts per million by mass (ppmw)
The first three water-content values are commonly used with gases, and the
second and fourth values are used with liquids. The first and third values are
convenient, as they are easily related (21 ppmv per lb/MMscf [18 mg/Sm
3
]). Use
of ppmw requires knowledge of the hydrocarbon molecular weight for conversion
to other concentrations. Unfortunately, the dew point is commonly used because
it makes practical sense. However, no simple conversion method exists between
dew point and the other concentration units.
This chapter provides ways to estimate the water content of gases and then
discusses the common processes used for its removal. The same basic processes
apply to drying liquids, but the liquid, which has a higher density, alters the
6.2 WATER CONTENT OF HYDROCARBONS
Determining the saturation water content of a gas (the dew point) is a standard
but complex problem in thermodynamics, and excellent discussions of phase
equilibria calculations are given in many thermodynamics texts (e.g., Smith
mercial grades of propane require lower water contents (see Chapter 1).
concentration should be 0.1 ppmv or less. Chapter 7 discusses hydro-
processes slightly. Chapter 10 discusses dehydration of liquids.
© 2006 by Taylor and Francis Group, LLC
Gas Dehydration 135
et al. [2001] and Prausnitz et al. [1999]). We first assume ideal behavior, and then
present a commonly used engineering correlation on the basis of experimental
measurements.
In any mixture, where both the gas and liquid phases are in equilibrium, each
component, i, in the mixture obeys the relationship
(6.1)
where x
i
is the mole fraction in the liquid phase, g
i
is the activity coefficient, P
Sat
is the saturation or vapor pressure, y
i
is the vapor phase concentration, j
i
the
vapor phase fugacity coefficient, and P the total pressure. The vapor pressure is
temperature dependent, whereas the activity and fugacity coefficients are tem-
perature, pressure, and composition dependent. If three of the four variables,
temperature, pressure, liquid composition, and vapor composition, are known,
the fourth variable can be calculated (flash calculation). The activity and fugacity
coefficients can be calculated by use of equations of state and empirical equations.
However, present-day equations of state have difficulty modeling the strong polar
nature of water in hydrocarbon−water mixtures; calculations from different sim-
ulators can give different results, especially in prediction of water content in the
liquid phase. If the mixture contains methanol or brine, the calculations become
more uncertain.
However, reasonably good estimates of the concentration of water in the
vapor phase in equilibrium with liquid water can be made at pressures below
500 psia (35 bar). If we make the good assumption of negligible hydrocarbon in
the liquid water phase, which, thus makes both x
i
and g
i
unity for water and
assume the gas phase to be ideal, which makes j
i
unity, we obtain
(6.2)
Equation 6.2 provides reasonably good values for gas-phase water content,
provided that the gas contains less than a few mol% of either CO
2
or H
2
S.
Figure 6.1(a) gives the water content for a hydrocarbon gas as a function of
for salinity if the water phase contains brine. The chart is widely used in the gas
industry and gives good results, provided the gas contains less than about 5 mol%
CO
2
plus H
2
S. The Engineering Data Book (2004b) provides an empirical method
to predict water concentrations when the gas contains higher concentrations of
the acid gases. Note that the dashed lines in Figure 6.1(a) represent the water phase
in equilibrium with either ice or gas hydrate, depending upon the gas composition
x P y P
i i i
Sat
i i
γ ϕ =
y
P
P
H O
H O
Sat
2
2
=
temperature and pressure. Figure 6.1(b) provides corrections for gas gravity and
A more accurate way to determine the water content is to use Figure 6.1.
Appendix B tabulates water vapor pressure as a function of temperature.
© 2006 by Taylor and Francis Group, LLC
136 Fundamentals of Natural Gas Processing
FIGURE 6.1(a) Water content of hydrocarbon gases as a function of temperature and
pressure.
80,000
20,000
60,000
40,000
10,000
8,000
4,000
6,000
2,000
1,000
800
600
400
200
100
80
60
10
40
20
8
4
6
2
1
260 220 180 140 100 80 60 40 20 0 −20 −40 300
Temperature, °F
(a)
L
b

w
a
t
e
r
/
M
M
s
c
f

o
f

w
e
t

g
a
s
1
4
.
7
p
s
i
a
2
5
5
0
1
0
0
2
0
0
3
0
0
4
0
0
5
0
0
6
0
0
8
0
0
1
0
0
0
1
5
0
0
2
0
0
0
3
0
0
0
4
0
0
0
5
0
0
0
6
0
0
0
1
0
0
0
0
80,000
20,000
60,000
40,000
10,000
8,000
4,000
6,000
2,000
1,000
800
600
400
200
100
80
60
10
40
20
8
4
6
2
1
−60
H
y
d
r
a
t
e

f
o
r
m
a
t
i
o
n

l
i
n
e
Dashed lines denote metastable
region. Slope of lines is
composition dependent and
water content may be lower than
shown on chart
© 2006 by Taylor and Francis Group, LLC
Gas Dehydration 137
and pressure. Actual gas-phase water contents can be lower than that obtained
from the chart.
FIGURE 6.1(b) Gas gravity and salinity corrections to water content in hydrocarbon
gases. (Adapted from Engineering Data Book, 2004b.)
Total solids in brine, wt %
(b)
1.0
0.9
0.8
0.7
0.6 0.8 1.0 1.2 1.4 1.6 1.8
C
G
Molecular weight
Correction for gas gravity
20 25 30 35 40 45 50
0.90
0 1 2 3 4
0.92
0.94
0.96
0.98
1.00
Correction for salinity
Gas gravity
3
0
0
°
F
2
5
0
°
F
2
0
0
°
F
1
5
0
°
F
1
0
0
°
F
5
0
°F
C
S
=
H
2
O

w
i
t
h

b
r
i
n
e
H
2
O

w
i
t
h
o
u
t

b
r
i
n
e
Example 6.1 Calculate the water content of the sweet natural gas shown in Table 6.1
at 300 psia (20.7 bar) and 80°F (26.7°C) by use of Equation 6.2 and Figure 6.1.
© 2006 by Taylor and Francis Group, LLC
138 Fundamentals of Natural Gas Processing
mol fraction
Next convert to lb Water/MMscf:
From Table 6.1, the MW of the gas mixture is 18.41 and the specific gravity is
SpGr = r
gas
/r
air
= 18.41/28.96 = 0.636
Sat
= 85 lb/MMscf (1,400 mg/Sm
3
). Correct for specific gravity
G G
W
Sat
= (0.99)(85) = 84 lb/MMscf (1,400 mg/Sm
3
)
The values differ by 4%. Increasing the pressure to 1,000 psia (69 bar). Equation
6.2 is 27% below the value obtained from Figure 6.1a.
6.3 GAS DEHYDRATION PROCESSES
This section discusses the conventional methods for drying natural gas, and then
briefly describes some less-conventional methods. Two processes, absorption and
adsorption, are the most common and are discussed in more detail.
TABLE 6.1
Gas Composition Data for Example 6.1
Component Mole Fraction Molar Mass Mol Fraction × Molar Mass
CH
4
0.90 16.043 14.44
C
2
H
6
0.04 30.070 1.20
C
3
H
8
0.03 44.097 1.32
n-C
4
H
10
0.02 58.123 1.16
N
2
0.01 28.013 0.28
Totals 1.00 18.41
y
P
P
psia
psia
H O
H O
Sat
2
2
0 507
300
0 0017 = = =
.
.
W =










0 0017 18 .
mol H O
mol gas
lbH O
lb mol
2 2
-








=
=
1
379 5
10 81
6
lb mol
scf
lbH O
MMscf
2
-
.
( )
11 300
3
,
mg
m






From Figure 6.1(a) W
by obtaining C from Figure 6.1b (C = 0.99), and multiplication gives
Equation 6.2 using vapor-pressure data from Appendix B gives
© 2006 by Taylor and Francis Group, LLC
Gas Dehydration 139
6.3.1 ABSORPTION PROCESSES
6.3.1.1 Overview of Absorption Process
Water levels in natural gas can be reduced to the 10 pmmv range in a physical
absorption process in which the gas is contacted with a liquid that preferentially
absorbs the water vapor. The solvent used for the absorption should have the
following properties:
• A high affinity for water and a low affinity for hydrocarbons
• A low volatility at the absorption temperature to reduce vaporization
losses
• A low viscosity for ease of pumping and good contact between the gas
and liquid phases
• A good thermal stability to prevent decomposition during regeneration
• A low potential for corrosion
In practice, the glycols, ethylene glycol (EG), diethylene glycol (DEG), trieth-
ylene glycol (TEG), tetraethylene glycol (TREG) and propylene glycol are the most
commonly used absorbents; triethylene glycol is the glycol of choice in most
instances. For operations in which frequent brine carryover into the contactor occurs,
operators use EG because it can hold more salt than the other glycols. The solubility
of sodium chloride in EG water mixtures is around 20 wt% (Masaoudi, 2004; Parrish,
2000; Trimble, 1931), whereas it is only around 5 wt% in TEG (Kruka, 2005).
glycols. Table 6.2 shows their formulas, along with the maximum recommended
TABLE 6.2
Glycols Used in Dehydration
Name Formula
Maximum
Recommended
Regeneration
Temperature, °F (°C)
a
Ethylene
glycol (EG)
HO—(CH
2
)
2
—OH C
2
H
6
O
2
Diethylene glycol
(DEG)
HO—((CH
2
)
2
—O)—(CH
2
)
2
—OH C
4
H
10
O
3
325 (160)
Triethylene glycol
(TEG)
HO—((CH
2
)
2
—O)
2
—(CH
2
)
2
— OH C
6
H
14
O
4
360 (180)
Tetraethylene glycol
(TREG)
HO—((CH
2
)
2
—O)
3
—(CH
2
)
2
— OH C
8
H
18
O
5
400 (200)
Propylene glycol HO—(CH
2
)
3
—OH C
3
H
8
O
2
a
Source: Kohl and Nielsen (1997).
Appendix B gives some of the more important properties of the commonly used
© 2006 by Taylor and Francis Group, LLC
140 Fundamentals of Natural Gas Processing
regeneration temperature. Smith (2004) reports temperatures for onset of slow
thermal degradation as 328°F (164°C) and 385°F (196°C) for DEG and TEG,
respectively. However, the author gives typical maximum regenerator temperatures
of about 350°F (175°C) and 400°F (204°C) for DEG and TEG, respectively.
Manufacturer’s literature (Dow Chemical, 2003) gives a decomposition temper-
ature of 464°F (240°C) for TEG.
Figure 6.2 shows a typical, simplified flow sheet for a glycol absorption unit.
The wet gas passes through an inlet scrubber to remove solids and free liquids,
and then enters the bottom of the glycol contactor. Gas flows upward in the
contactor, while lean glycol solution (glycol with little or no water) flows down
over the trays. Rich glycol absorbs water and leaves at the bottom of the column
while dry gas exits at the top. The rich glycol flows through a heat exchanger at
the top of the still where it is heated and provides the coolant for the still
condenser. Then the warm solution goes to a flash tank, where dissolved gas is
removed. The rich glycol from the flash tank is further heated by heat exchange
with the still bottoms, and then becomes the feed to the still. The still produces
water at the top and a lean glycol at the bottom, which goes to a surge tank before
being returned to the contactor.
Operating conditions for glycol units are governed principally by the degree
of dehydration required, the physical properties of the glycol solutions, and the
inlet pressure of the gas to the processing unit. Some typical operating conditions
for TEG absorbers are given in the next section. The articles of Hernandez-
Valencia et al (1992), Parrish et al. (1986), and Wieninger (1991), along with the
FIGURE 6.2 Schematic of typical glycol dehydrator unit. (Adapted from Engineering
Data Book, 2004b.)
Surge drum
Filter
Wet inlet
gas
Glycol
dehydrator
Liquids
Dry gas
Lean
glycol
Flash gas
Rich
glycol
Water vapor
F
l
a
s
h

t
a
n
k
Inlet
scrubber
Steam or
hot oil
© 2006 by Taylor and Francis Group, LLC
Gas Dehydration 141
Engineering Data Book (2004b), provide detailed information on absorber design
and operation. Although many design references refer to trays, most modern TEG
absorbers use structured packing instead of trays (McCartney, 2005).
6.3.1.2 Representative Operating Conditions
for TEG Absorbers
As noted above, triethylene glycol is the most common absorbent for dehydration,
although some operations in colder climates (e.g., the North Sea) will use dieth-
ylene glycol because of its lower viscosity. Tetraethylene glycol has a lower vapor
pressure and withstands higher regeneration temperatures than does TEG, but the
additional cost outweighs the marginal increased benefits. Smith (2004) reports
about 20,000 TEG dehydration units operating in the United States.
Gas flow rates, compositions, inlet pressures and temperatures, and the
useful guidelines for typical service.
solutions of different concentrations as a function of contactor temperature. For
example, if an equilibrium dew point of −15°F (−26°C) is required and the
contactor operates at 80°F (27°C), then a TEG solution of 99.5 wt% is required.
However, the assumption is that the dry gas exiting is in equilibrium with the
incoming lean glycol (i.e., infinite number of contactor trays). To account for
nonequilibrium concentrations, the Engineering Data Book (2004b) suggests use
of an equilibrium temperature that is 10 to 20°F (5 to 10°C) below the desired
dew point temperature. Therefore, to obtain the −15°F (−26°C) exit dew-point
temperature, the lean glycol concentration should be 99.8 to 99.9 wt%.
The Engineering Data Book (2004b) notes that Figure 6.3 is based upon
equilibrium between water vapor and a liquid water phase. At lower temperatures,
the true equilibrium condensed phase is gas hydrate, which will form at higher
temperatures than does subcooled water. Therefore, solids formation could be as
much as 15 to 20°F (8 to 10°C) higher than the dew point value obtained from
Figure 6.3. The actual error depends upon temperature, pressure, and gas com-
position. However, the pressure effect is minimal, and the chart can be used to
1,500 psia (100 bara).
The required lean-glycol concentration dictates still reboiler operating con-
ditions. Higher reboiler temperatures yield leaner glycol. At a 400°F (204°C),
the typical maximum regeneration temperature, TEG yields a lean-glycol con-
centration of 98.6 wt% (Engineering Data Book, 2004b) at sea level. Higher
purity requires reduction of the partial pressure of water in the reboiler vapor
space. The most common way to achieve this pressure reduction is to use a
stripping gas* or vacuum distillation, which yields lean glycol concentrations
of 99.95 and 99.98, respectively. The Engineering Data Book (2004b), Smith
* A stripping gas is a noncondensing gas at operating conditions that is used to strip the more volatile
solutes from a liquid.
required degree of dehydration vary widely. However, Table 6.3 provides some
Figure 6.3 shows the equilibrium dew points that can be achieved with TEG
© 2006 by Taylor and Francis Group, LLC
142 Fundamentals of Natural Gas Processing
(1993), and Smith and Humphrey (1995) describe these methods, along with two
proprietary methods, one of which claims to obtain a 99.999 wt% glycol.
As a general guideline, absorber inlet temperatures can be as high as 150°F
(66°C), although temperatures above 100°F (38°C) may result in unacceptable
vaporization losses for the glycol solutions (Engineering Data Book, 2004b).
At inlet temperatures below approximately 50°F (10°C), the viscosities of the
cies and increase pumping costs. Typical maximum operating pressures are
around 2,000 psig (140 barg).
TABLE 6.3
Typical Operating Conditions of TEG Dehydrators
Contactor
Inlet pressures < 2000 psig (139 bar)
Inlet temperatures 60°F to 100°F (16°C to 38°C)
(Lower temperatures enhance absorption capacity but
can lead to hydrate formation at high pressure.)
Pressure drop 5 to 10 psi (34 to 69 kPa)
Glycol circulation rate 2 to 5 gal/lb H
2
O removed, with 3 common.
(17 to 42 L/kg)
Tray efficiencies 25 to 30%.
Dew points > −25°F (−32°C) (Enhanced regeneration required for lower dew
point temperatures.)
Glycol losses
Vaporization ∼ 0.012 gal/MMscf (1.6 L/MM Sm
3
)
Total 0.025 gal/MMscf (3.3 L/MM Sm
3
)
Regenerator
(reboiler and still)
Column internals Packed equivalent to 3 or 4 trays
Reboiler temperatures 375°F to 400°F (191°C to 204°C)
Flash tank
Pressure 50 to 75 psig (446 to 618 kPa)
Temperature 150°F (66°C)
Retention times
a
C4+ Lean Gas
C4+ Rich Gas
a

∼10 minutes.
∼20 minutes (Use three-phase separator.) TEG absorbs about 1 scf
gas/gal TEG at 1,000 psig and 100°F (0.0076 Sm
3
/L at 70 barg
and 38°C)
a
For treatment of gas streams that contain high concentrations of C4+, adequate time needs
to be allowed for removal of the less-volatile components from the glycol to minimize hydro-
carbon losses in the still overhead.
Source: Engineering Data Book, (2004b).
Figure 6.3 shows that lower absorption temperatures improve dehydration.
glycol solutions (see Appendix B) are sufficiently high to reduce plate efficien-
© 2006 by Taylor and Francis Group, LLC
Gas Dehydration 143
Triethylene glycol losses from a properly operating plant, excluding spill-
age, should be minimal. The Engineering Data Book (2004b) estimates about
0.1 gal/MMscf (13 L/10
6
Sm
3
) from carryover if a standard mist eliminator is used.
Other losses range between 0.05 gal/MMscf (7 L/MMSm
3
) for high-pressure,
FIGURE 6.3 Equilibrium water dew point as a function of contactor temperature and
TEG concentration in wt%. (Adapted from Engineering Data Book. Used with permission
of Fluor Enterprises.)
40 100 80 60 120 140 160 180
−120
−100
−80
−60
−40
−20
40
0
20
60
Contactor temperature, °F
E
q
u
i
l
i
b
r
i
u
m

d
e
w

p
o
i
n
t

t
e
m
p
e
r
a
t
u
r
e
,

°
F
90 95 97 98 99 99.5
99.8
99.9
99.95
99.97
99.98
99.99
99.995
99.997
99.999
© 2006 by Taylor and Francis Group, LLC
144 Fundamentals of Natural Gas Processing
low-temperature dehydration and 0.3 gal/MMscf (40 L/MMSm
3
) for low-pressure,
high-temperature dehydration. Higher losses suggest foaming problems in the
absorber or contactor.
If the dehydrator is used to dry high-pressure CO
2
streams, the glycol losses
can be much higher because glycol is more soluble in dense CO
2
than in natural
gas. The Engineering Data Book (2004b) refers to the paper by Wallace (1985)
on this subject.
Sour gases can be dried with glycol solutions if appropriate anticorrosion
TEG as a function of temperature and pressure in lean TEG. These figures are used
to estimate the vapor volumes, and this is added to the estimated 1 scf hydrocarbon
gas/gal TEG at 1,000 psig and 100°F (0.0076 Sm
3
/L at 70 barg and 38°C).
6.3.1.3 Other Factors That Affect Glycol Dehydrator
Performance
Oxygen reacts with the glycols to form corrosive acidic compounds. The products
also increase the potential for foaming and glycol carryover (Wieninger, 1991).
A dry natural gas blanket is often put over the storage and surge tanks to minimize
air intrusion. The Engineering Data Book (2004b) suggests that precautions be
taken if oxygen is present in the gas but offers no options.
FIGURE 6.4 Solubility of CO
2
in lean triethylene glycol as a function of temperature and
partial pressure (Wichert, 2005).
60
800
Partial pressure
psia
700
600
500
400
200
300
150
125
100
80
60
40
25
15
S
c
f

C
O
2
/
g
a
l
l
o
n

T
E
G
160 140 120 100 80
0.1
180
1.0
10.0
Temperature, °F
measures are taken. Figures 6.4 and 6.5 show the solubility of the acid gases in
© 2006 by Taylor and Francis Group, LLC
Gas Dehydration 145
The Engineering Data Book (2004b) notes that low pH accelerates glycol
decomposition. The book suggests addition of trace amounts of basic hydrocarbons
(e.g., alkanolamines). For glycol units downstream of amine treaters, the trace
amount of amine carryover tends to mitigate the acidity problem.
Glycol dehydration units in the typical gas plant are downstream of the amine
treaters; consequently, they process a relatively clean gas stream that contains no
H
2
S and reduced CO
2
levels. Amine carryover is the main concern in these
situations. Dehydrators in gathering systems can be in more demanding situations
where they may process streams that contain acid gases and methanol or other
hydrate inhibitors. They also may be subjected to oil and liquid water slugs, as
well as compressor oil mist. Water-soluble components increase the regenerator
duty, and hydrocarbon impurities can cause foaming in the system. Proper sizing
of inlet scrubbers is critical in these situations. In some cases, filter separators or
coalescing filters are required to prevent fine mists from entering the absorber.
As noted above, offshore producers sometimes use (EG) instead of TEG when
produced water is carried over to the contactor. In these situations, the purpose of
dehydration is hydrate prevention. To meet the 4- to 7-lb H
2
O/MMscf (60 to 110
mg/Sm
3
) specification, either stripping gas or running the still at a vacuum is
required. The higher (EG) evaporation losses are considered acceptable. In some
cases, reclaimers, which distill the (EG) from the brine, are on the platform.
The Engineering Data Book (2004b) reports that glycol removes 40 to 60%
of methanol from the feed. The increased liquid loading places an additional heat
FIGURE 6.5 Solubility of H
2
S in lean triethylene glycol as a function of temperature and
partial pressure (Wichert, 2005).
400
Partial pressure
psia
300
200
150
100
80
60
40
25
10
7
125
160 140 120 100 80 180
Temperature, °F
60
S
c
f


H
2
S
/
g
a
l
l
o
n

T
E
G

0.1
1.0
10.0
100.0
© 2006 by Taylor and Francis Group, LLC
146 Fundamentals of Natural Gas Processing
load on the reboiler and an added vapor load on the regenerator. Even with adverse
conditions, many unmanned glycol systems function well in the field.
6.3.2 ADSORPTION PROCESSES
6.3.2.1 Overview of Adsorption
The two types of adsorption are physical adsorption and chemisorption. In phys-
ical adsorption, the bonding between the adsorbed species and the solid phase is
called van der Waals forces, the attractive and repulsive forces that hold liquids
and solids together and give them their structure. In chemisorption, a much
stronger chemical bonding occurs between the surface and the adsorbed mole-
cules. This chapter considers only physical adsorption, and all references to
adsorption mean physical adsorption.
Physical adsorption is an equilibrium process like vapor−liquid equilibria and
equations analogous to Equation 6.1 apply. Thus, for a given vapor-phase con-
centration (partial pressure) and temperature, an equilibrium concentration exists
on the adsorbent surface that is the maximum concentration of the condensed
tions for water on a commercial molecular sieve. Such curves are called isotherms.
The figure is based upon a water−air mixture but is applicable to natural gas
systems. The important parameter is the partial pressure of water; total pressure
has only a minor effect on the adsorption equilibrium.
Because adsorbate concentrations are usually low, generally only a few layers
of molecules will build up on the surface. Thus, adsorption processes use solids
with extremely high surface-to-volume ratios. Commercially used synthetic
zeolites (i.e, molecular sieves) have surface-to-volume ratios in the range of
750 cm
2
/cm
3
, with most of the surface for adsorption inside of the adsorbent. In
the case of molecular sieves, the adsorbent consists of extremely fine zeolite
particles held together by a binder. Therefore, adsorbing species travel through
the macropores of the binder into the micropores of the zeolite. Adsorbents such
as silica gel and alumina are formed in larger particles and require no binder.
Pore openings that lead to the inside of commercial adsorbents are of molecular
size; they normally range from approximately 4 Å (1 Å = 10
−8
cm) to 100 Å.
Molecular sieves have an extremely narrow pore distribution, whereas silica gel
and alumina have wide distributions. However, a molecular sieve binder, which
is usually about 20% of the weight of the total adsorbent, has large pores capable
of adsorbing heavier components.
Two steps are involved in adsorbing a trace gas component. The first step is
to have the component contact the surface and the second step is to have it travel
through the pathways inside the adsorbent. Because this process is a two-step
process and the second step is relatively slow, solid adsorbents take longer to
come to equilibrium with the gas phase than in absorption processes.
In addition to concentration (i.e., partial pressure for gases), two properties
of the adsorbate dictate its concentration on the absorbent surface: polarity and
size. Unless the adsorbent is nonpolar, which is not the case for those used in
component (adsorbate) on the surface. Figure 6.6 shows the equilibrium condi-
G
a
s

D
e
h
y
d
r
a
t
i
o
n
1
4
7
FIGURE 6.6 Water loading on UOP Adsorbent 4A-DG MOLSIV Pellets. Activation conditions for the adsorbent were
662°F (350°C) and less than 10 microns Hg. (Adapted from Engineering Data Book 2004 b. Used with permission of UOP
LLC.)
1.0 E−8 1.0 E−7 1.0 E−6 1.0 E−5 1.0 E−4 1.0 E−3 1.0 E−2 1.0 E−1 1.0 E0 1.0 E+1 1.0 E+2 1.0 E+3 1.0 E+4
5
0
10
15
20
25
600°F
500°F
400°F
150°F
100°F
77°F
32°F
Partial pressure of water, psia
L
b

w
a
t
e
r

a
d
s
o
r
b
e
d
/
1
0
0

l
b

a
c
t
i
v
a
t
e
d

a
d
s
o
r
b
e
n
t
© 2006 by Taylor and Francis Group, LLC
© 2006 by Taylor and Francis Group, LLC
148 Fundamentals of Natural Gas Processing
gas plants, polar molecules, like water, will be more strongly adsorbed than
weakly polar or nonpolar compounds. Thus, methane is displaced by the weakly
polar acid gases that are displaced by the strongly polar water.
How size affects adsorption depends upon the pore size of the adsorbent. An
adsorbate too large to fit into the pores adsorbs only on the outer surface of
adsorbent, which is a trivial amount of surface area compared with the pore area.
If the pores are sufficiently large to hold different adsorbates, the less volatile,
which usually correlates with size, adsorbates will displace the more volatile
ones. Therefore, ethane is displaced by propane.
In commercial practice, adsorption is carried out in a vertical, fixed bed of
adsorbent, with the feed gas flowing down through the bed. As noted above, the
process is not instantaneous, which leads to the formation of a mass transfer zone
(MTZ) in the bed. Figure 6.7 shows the three zones in an adsorbent bed:
1. The equilibrium zone, where the adsorbate on the adsorbent is in
equilibrium with the adsorbate in the inlet gas phase and no additional
adsorption occurs
2. The mass transfer zone (MTZ), the volume where mass transfer and
adsorption take place
3. The active zone, where no adsorption has yet taken place
In the mass transfer zone (MTZ), the concentration drops from the inlet value,
y
in
, to the outlet value, y
out
, in a smooth S-shaped curve. If the mass transfer rate
were infinite, the MTZ would have zero thickness. The MTZ is usually assumed
to form quickly in the adsorption bed and to have a constant length as it moves
FIGURE 6.7 Vapor-phase concentration profile of an adsorbate in the three zones of an
adsorption bed.
Bed length
V
a
p
o
r

p
h
a
s
e


c
o
n
c
e
n
t
r
a
t
i
o
n y
In
0
y
Out
Equilibrium
zone
Mass transfer
zone
Active
zone
L
© 2006 by Taylor and Francis Group, LLC
Gas Dehydration 149
through the bed, unless particle size or shape is changed. The value of y
in
is
dictated by upstream processes; the y
out
value is determined by the regeneration gas
adsorbate content.
The length of the MTZ is usually 0.5 to 6 ft (0.2 to 1.8 m), and the gas is in
the zone for 0.5 to 2 seconds (Trent, 2004). To maximize bed capacity, the MTZ
needs to be as small as possible because the zone nominally holds only 50% of
the adsorbate held by a comparable length of adsorbent at equilibrium. Both tall,
slender beds, which reduce the percentage of the bed in the MTZ, and smaller
particles make more of the bed effective. However, smaller particle size, deeper
beds, and increased gas velocity will increase pressure drop.
For a point in the MTZ, the gas phase adsorbate content increases in time
from y
in
to y
out
In principle, beds can be run until the first sign of breakthrough. This practice
maximizes cycle time, which extends bed life because temperature cycling is a
major source of bed degeneration, and minimizes regeneration costs. However,
most plants operate on a set time cycle to ensure no adsorbate breakthrough.
Trent (2004) presents data that show a change in the L/D from 0.8 to 2.7 in
the bed increases the useful adsorption capacity from 8.7 to 10.0 wt% in useful
water capacity for an equal amount of gas dried. However, the pressure drop
increases from 0.4 to 4.3 psi (0.020 to 0.20 kPa).
When used as a purification process, adsorption has two major disadvantages:
• It is a fixed-bed process that requires two or more adsorption beds for
continuous operation.
• It has limited capacity and is usually impractical for removing large
amounts of impurity.
However, adsorption is very effective in the dehydration of natural gas
because water is much more strongly adsorbed than any of the alkanes, carbon
dioxide, or hydrogen sulfide. Generally, a higher degree of dehydration can be
achieved with adsorbents than with absorption processes.
6.3.2.2 Properties of Industrial Adsorbents for Dehydration
Three types of commercial adsorbents are in common use in gas processing plants:
• Silica gel, which is made of pure SiO
2
• Activated alumina, which is made of Al
2
O
3
• Molecular sieves, which are made of alkali aluminosilicates and can
be altered to affect adsorption characteristics
primarily from commercial literature. The properties are representative and vary
between manufacturers.
Table 6.4 lists the more important properties of three adsorbents compiled
in an S-shaped curve that mirrors the curve shown in Figure 6.7.
© 2006 by Taylor and Francis Group, LLC
150 Fundamentals of Natural Gas Processing
Silica gels are used mostly where a high concentration of water (>1 mol%)
vapor is present in the feed, and low levels of water in the dehydrated gas are
not needed. They are relatively noncatalytic compounds. Aluminas are very polar
and strongly attract water and acid gases. They are used for moderate levels of
water in the feed when low levels of water in the product are not required. They
have the highest mechanical strength of the adsorbents considered here. How-
ever, for gas going into cryogenic processing, the only adsorbent that can obtain
the required dehydration is a molecular sieve. Of these, 4A is the most common,
but the smaller pore 3A is sometimes used. It has the advantage of being a poorer
catalyst for generation of COS if both H
2
S and CO
2
are present* because a
TABLE 6.4
Representative Properties of Commercial Silica Gels, Activated Alumina,
and Molecular Sieve 4A
Silica Gel Activated Alumina Molecular Sieve 4A
Shape Spherical Spherical Pellets (extruded
cylinders) and beads
Bulk density
lb/ft
3
(kg/m
3
)
49 (785) 48 (769) 40 −45 (640 − 720)
Particle size 4 − 8 mesh
5 −2 mm
7−14 mesh, 1/8-inch,
3/16-inch, 1/4-inch
diameter (3-mm,
5-mm, 6-mm)
1/16-inch,1/8-inch,1/4-inch
diameter cylinders
(1.6-mm, 3.2-mm, 6-mm)
Packed bed % voids 35 35 35
Specific heat
Btu/lb-°F (kJ/kg-K)
0.25 (1.05) 0.24 (1.00) 0.24 (1.00)
Surface area m
2
/g 650 − 750 325 − 360 600 − 800
Pore volume cm
3
/g 0.36 0.5 0.28
Regeneration
temperature, °F (°C)
375 (190) 320 to 430 (160 to 220) 400 to 600 (200 to 315)
Average pore
diameter (Å)
22 NA 3,4,5,10
Minimum dew
point temperature
of effluent, °F (°C)
a
−80 (−60) −100 (−75) −150 (−100)
Average minimum
moisture content of
effluent gas, ppmv
5 −10 10 − 20 0.1
a
As reported by Blachman and McHugh (2000).
* Carbonyl sulfide (COS) is formed in the following reaction: H
2
S + CO
2

↔ H
2
O + COS. The
equilibrium constant for the reaction is of the order of magnitude of 10
−6
at adsorption temperatures
but increases to 10
−4
at regeneration temperatures (Trent et al., 1993). Its concentration in feed gas
is normally extremely low.
© 2006 by Taylor and Francis Group, LLC
Gas Dehydration 151
portion of the more active sodium cations in 4A has been replaced with potas-
sium. If both oxygen and H
2
S are present 3A reduces the production of elemental
sulfur that can block adsorbent pores. However, plant operators usually have
little incentive to use 3A for dehydrating gas going to hydrocarbon recovery.
Blachman and McHugh (2000) discuss use of multiple adsorbents in the same
bed for applications in which both higher water concentrations and acid gases
are present.
6.3.2.3 Adsorption Process
Although this discussion uses molecular sieve as the example of an adsorbent
to remove water, with the exception of regeneration temperatures, the basic
process is the same for all gas adsorption processes. Figure 6.8 shows a sche-
matic of a two-bed adsorber system. One bed, adsorber #1 in Figure 6.8, dries
gas while the other bed, adsorber #2, goes through a regeneration cycle. The
wet feed goes through an inlet separator that will catch any entrained liquids
before the gas enters the top of the active bed. Flow is top-down to avoid bed
fluidization. The dried gas then goes through a dust filter that will catch fines
FIGURE 6.8 Schematic of a two-bed adsorption unit. Valving is set to have absorber #1
in drying cycle and absorber #2 in regeneration cycle. (Adapted from Engineering Data
Book, 2004b.)
A
d
s
o
r
b
e
r

#
1
A
d
s
o
r
b
e
r

#
2
Wet gas
Dry gas
Condenser
Water
knockout
Water
Inlet
separator
Open Closed
Compressor
Regeneration gas recycle
Regeneration gas
heater
Dust
filter
© 2006 by Taylor and Francis Group, LLC
152 Fundamentals of Natural Gas Processing
before the gas exits the unit. This filter must be kept working properly, especially
if the gas goes on to a cryogenic section with plate-fin heat exchangers, as dust
can collect in the exchangers and reduce heat transfer and dramatically increase
pressure drop.
regenerated. (Sales gas is sometimes used instead of a slip stream. The sales gas
stream has the advantage of being free of heavier hydrocarbons that can cause
coking.) This gas is usually about 5 to 10% of gas throughput. Regeneration
involves heating the bed, removing the water, and cooling. For the first two
steps, the regeneration gas is heated to about 600°F (315°C) to both heat the
bed and remove adsorbed water from the adsorbent. If COS formation is a
problem, it can be mitigated by lowering regeneration temperatures to 400 to
450°F (200 to 230°C) or lower, provided sufficient time for regeneration is
available, or by switching to 3A. Regeneration gas enters at the bottom of the
bed (countercurrent to flow during adsorption) to ensure that the lower part of
the bed is the driest and that any contaminants trapped in the upper section of
the bed stay out of the lower section. The high temperature required makes this
step energy intensive and in addition to furnaces, other heat sources (e.g., waste
heat from gas turbines that drive compressors) are used when possible. The hot,
wet regeneration gas then goes through a cooler and inlet separator to remove
the water before being recompressed and mixed with incoming wet feed. To
complete the regeneration, unheated regeneration gas passes through the bed to
cool before it is placed in drying service. Gas flow during this step can be
concurrent or countercurrent.
The Engineering Data Book (2004b) recommends that the bed pressure not
be changed more than 50 psi/min (6 kPa/s). Therefore, if the adsorption process
operates at high pressure, regeneration should take place at as high a pressure as
possible to reduce the time needed for changing the pressure. However, as Malino
(2004) points out, higher pressures increase the amount of water and hydrocarbons
that condense at the top of the bed and fall back onto the adsorption bed. This
unavoidable refluxing is a major cause of bed aging, as it leads to adsorbent
breakdown and subsequent fines agglomeration (Richman, 2005). The caking
leads to higher pressure drop. Condensation at the bed walls can also occur, which
can cause bed channeling.
dehydrators (Chi and Lee, 1973; Cummings, 1977; Engineering Data Book,
2004b; Kohl and Nielsen,1997; Lukchis [date unknown]; Petty, 1976; UOP,1991;
Trent, 2001). Both the Engineering Data Book (2004b) and Trent (2001) give
extensive details on designing adsorber systems.
The combination of feed rate, pressure drop, and adsorbent crush strength
dictates the adsorption bed geometry. As noted in the above discussion regarding
minimizing MTZ thickness, the bed diameter should be kept small. This feature
also reduces the wall thickness of the high-pressure vessels and increases the
superficial velocity, which improves mass transfer in the gas phase. However, it
does not affect intraparticle mass transfer, which is the slower of the two processes.
Table 6.5 lists design parameters that are guidelines for typical molecular sieve
Figure 6.8 shows a slip stream of dry gas returning to the bed that is being
© 2006 by Taylor and Francis Group, LLC
Gas Dehydration 153
Thus, higher velocities increase the MTZ thickness. Accurate calculation of the
MTZ thickness is complex. Trent (2004) suggests the following dimensioned
equation for estimating the thickness of the MTZ, L
MTZ
, in feet of 4 × 8 (1/8-inch
[3 mm] diameter) mesh beads:
L
MTZ
(ft) = 2.5 + 0.025 V
S
(ft/min), (6.3)
where V
S
is the superficial gas velocity in ft/min. For 8 × 12-mesh (1/16-inch)
beads, the length is about 70% of the values calculated by Equation 6.3. (Multiply
m/s by 197 to obtain ft/min and feet by 0.30 to obtain meters.) The Engineering
Data Book (2004b) suggests the following equation:
L
MTZ
(ft) = F [V
S
(ft/min)/35]
0.3
, (6.4)
where the factor F = 1.70 ft for a 1/8-inch (3-mm) sieve and 0.85 for 1/16-inch
(1.5-mm) material. Over the typical gas flow ranges, Equation 6.3 gives an L
MTZ
double that of Equation 6.4.
As noted above, higher velocities increase pressure drop through the bed.
This pressure drop has two adverse effects:
• Higher inlet compression discharge pressures to maintain the same
refrigeration requirements and outlet pressure
• Increased mechanical load on the adsorbent, which leads to particle
breakdown and causes further increases in pressure drop
The adsorbent beds typically have a 6-inch (15-cm) deep layer of inert 1/2 -
to 1-inch (13- to 25-mm) diameter alumina or ceramic balls (density of about
TABLE 6.5
Typical Operating Conditions for Molecular Sieve Dehydration Units
Feed rate 10 to 1500 MMscfd (0.3 to 42 MMSm
3
/d)
Superficial velocity Approximately 30 to 35 ft/min (9 to 11 m/min)
Pressure drop Approximately 5 psi (35 kPa), not to exceed 10 psi (69 kPa)
Cycle time Four to 24 hours; 8 or a multiple thereof is common
Temperatures and pressures
Adsorption
Regeneration
Temperatures: 50 to 115°F (10 to 45°C)
Pressures: to 1500 psig (100 barg),
Temperatures: 400 to 600°F (200 to 315°C)
Pressures: Adsorption pressure or lower.
Source: See text for references.
© 2006 by Taylor and Francis Group, LLC
154 Fundamentals of Natural Gas Processing
80 lb/ft
3
[1,200 kg/m
3
]) resting on a floating screen at the top of the bed. This
layer is on top of another floating screen-supported layer of 1/8-inch (3 mm)
diameter beads. These layers help distribute the incoming gas flow but, more
importantly, help keep the bed from shifting. The screens float to account for
thermal expansion during regeneration. The bottom of the bed has a similar
layering, with the smaller beads on top of the larger ones, which are supported
by a fixed screen. Some manufacturers offer molecular sieves that are more
resistant to the attrition caused by refluxing at the top of the bed (Richman, 2005).
The most common reasons for replacing a bed are loss of adsorbent capacity
and unacceptable pressure drop, which usually occur simultaneously. Values for
the loss of capacity with time vary considerably, but common values used for
molecular sieves in dehydration service are a 35% capacity loss over a 3 to 5 year
period or a 50% loss in approximately 1,600 cycles. Typically, a rapid loss occurs
in the beginning and a gradual loss thereafter. The adsorbent decays primarily
because of carbon and sulfur fouling and caking caused by instability in the clay
binder. These effects occur during bed regeneration. de Bruijn et al. (2001) provides
a complete discussion of these phenomena.
Increased pressure drop is usually caused by breakdown of adsorbent into
finer particles and by caking at the top of the bed because of refluxing. Attrition
can occur when the pressure is increased or decreased after or before regeneration.
Monitoring the pressure drop is important, as it provides a good diagnostic to
bed health. The Engineering Data Book (2004b) recommends a modified form
of the Ergun equation to compute pressure drop:
DP/L (psi/ft) = BµV
S
+ CrV
S
2
, (6.5)
with viscosity, m, in centipoises, density, r, in lb/ft
3
, and superficial velocity, V
S
,
in ft/min (multiply psi/ft by 22.62 to obtain kPa/m and m/s by 197 to obtain
ft/min). The coefficient values for typical adsorbents are given in Table 6.6.
A number of factors affect bed capacity. For the commonly used 4A molec-
ular sieve, the Engineering Data Book (2004b) suggests that the design water
content of a molecular sieve when at equilibrium with saturated gas at 75°F
TABLE 6.6
Coefficient Values for Typical Adsorbents
Particle Type B C
1/8-inch (3-mm) bead 0.0560 0.0000889
1/8-inch (3-mm) pellets 0.0722 0.000124
1/16-inch (1.5-mm) bead 0.152 0.000136
1/16-inch (1.5-mm) pellets 0.238 0.000210
© 2006 by Taylor and Francis Group, LLC
Gas Dehydration 155
(24°C) will be 13 lb H
2
O/100 lb sieve compared with a new molecular sieve,
which holds about 20 lb H
2
O/100 lb sieve. Two factors affect this number:
water content of entering gas and adsorption temperature. Figure 6.9 (Engi-
neering Data Book 2004b). Provides a correction for the capacity as a function
a correction for the decrease in bed capacity at temperatures greater than 75°F
(24°C). The effect of unsaturated gas and higher temperature can be computed
by
C
SS
= 0.636 + 0.0826 ln(Sat) (6.6)
and
C
T
= 1.20 – 0.0026 t(°F) (6.7a)
C
T
= 1.11 – 0.0047 t(°C), (6.7b)
where C
SS
and C
T
are correction factors for subsaturation and temperature, respec-
tively. Sat is the percent of saturation.
Example 6.2 An existing 4A molecular sieve bed has been processing 80
MMscfd on a 12-hour cycle with two beds. Exit gas goes to a cryogenic
turboexpander section. Gas flow is increased to 100 Mscfd. Estimate the
increased pressure drop and determine whether the bed capacity allows contin-
ued operation on a 12-hour cycle or the cycle time should be changed. The gas
enters the bed at 120°F and 950 psig. Water content is 60% of saturation at
120°F. The molar mass of the gas is 18.5, with a viscosity of 0.014 cP and a
compressibility factor of 0.84.
FIGURE 6.9 Molecular-sieve capacity correction for unsaturated inlet gas. (Adapted from
Engineering Data Book, 2004b.)
of percent relative saturation. Figure 6.10 (Engineering Data Book 2004b) gives
© 2006 by Taylor and Francis Group, LLC
156 Fundamentals of Natural Gas Processing
The adsorption bed contains 41,000 lbs of 1/8-inch diameter beads with a bulk
density of 44 lb/ft
3
. The inside wall diameter of the bed is 7.5 ft. The absorbent
was installed 2 years ago.
Pressure-Drop Calculation
Determine the height of the adsorption bed. The volume of adsorbent V =
mass/adsorbent density = 41,000/44 = 932 ft
3
. The cross-sectional area of the bed
A = pD
2
/4 = p(7.5)
2
/4 = 44.2 ft
2
. The bed height then is V/A = 932/44.2 = 21.1 ft.
Gas density,
r = MW P/(zRT) = 18.5 × (950 + 14.7)/[ 0.84 × 10.73 (460 + 120)] = 3.41 lb/ft
3
.
To obtain superficial velocity, we first need the increased actual volumetric flow
rate, Q (Equation 4.18a):
The superficial velocity is then V
S
= Q/A = 992/44.2 = 22.4 ft/min.
FIGURE 6.10 Correction for decrease in adsorption capacity caused by inlet-gas
temperature in molecular sieve. (Adapted from Engineering Data Book, 2004b.)
0.70
0.75
0.80
0.85
0.90
0.95
1.00
70 80 90 100 110 120 130 140 150 160 170 180 190
Temperature, °F
C
o
r
r
e
c
t
i
o
n

f
a
c
t
o
r
,

C
T
Q scfm
P psia
T R z
z
R
=






°






14 7
520
1
1 1
.
( )
( )






=
×
+






100 10
1440
14 7
950 14 7
460
6
.
.
++












=
.
.
120
520
0 84
1
992 ft /min
3
© 2006 by Taylor and Francis Group, LLC
Gas Dehydration 157
Use of Equation 6.5, ∆P/L (psi/ft) = BµV
S
+ CrV
S
2
, with B = 0.0560 and C =
0.0000889 gives
∆P/L = (0.0560)(0.014)(22.4) + (0.000089)(3.41)(22.4)
2
= 0.170 psi/ft.
Total bed pressure-drop is 0.170 psi/ft × 21.1 ft = 3.58 psi, which is in a good
operating range.
Bed Capacity Calculation
To determine the capacity, first calculate the capacity per pound of absorbent. The
bed has aged, so at 75°F the absorbent should hold 13 lb water/100 lb of sieve if
the entering gas is saturated with water vapor. However, the gas enters at 60% of
or Equation 6.7 is to reduce capacity to 88%, so the bed should hold 12.6 × 0.88
= 11.1 lb water/100 lb sieve in the equilibrium zone.
The length of the MTZ is calculated by either Equation 6.3 or Equation 6.4, but
use the former because it is more conservative
L
MTZ
(ft) = 2.5 + (0.025 × 22.4) = 3.1 ft.
This value represents 15% (3/21) of the bed height. Assume the MTZ holds 50%
of the equilibrium loading; the bed should hold
11.1/100 × 41,000(0.85 + 0.15 × 0.5) = 4,210 lb of water.
is 100 lb/MMscf. Essentially, all water into the bed must be removed. The gas
enters at 60% of saturation, which is 0.6 × 100 = 60 lb/MMscf. For a 12-hour
cycle, the water adsorbed is 60 × 100 × 0.5 = 3,000 lb water. The bed is slightly
oversized but can remain on a 12-hour cycle.
The major operating costs of adsorption are the energy required for regeneration
and the compression power required to overcome bed pressure drop. To minimize
the heat load, the adsorption beds are insulated. Insulation may be external or internal.
Internal insulation saves energy during bed regeneration because it eliminates heating
of the vessel walls and reduces the regeneration time. However, insulation imper-
fections and cracks may cause wet gas to bypass the adsorbent. Internal insulation
also requires a larger diameter pressure vessel, which adds to capital costs.
Regeneration usually takes 8 or more hours with 50 to 60% of the time
involved in heating and driving off water. The balance of time is in cooling down
the bed and having about 30 minutes of standby and switching.
The Engineering Data Book (2004b) recommends the following simple and
conservative technique for estimating the heat required for regeneration, q.
q
W
= (1800 Btu/lb)(lbs of water on the bed) (6.8a)
q
W
= (4200 kJ/kg)(kg of water on the bed) (6.8b)
saturation. With either Figure 6.9 or Equation 6.6, the absorbent holds only 97%
of capacity or 12.6 lb water/100 lb sieve. The temperature effect from Figure 6.10
From Figure 6.1a, the water content at saturation for a gas at 950 psig and 120°F
© 2006 by Taylor and Francis Group, LLC
158 Fundamentals of Natural Gas Processing
q
si
= (lb of sieve)(0.24 Btu/lb °F)(T
rg
− T
1
) (6.9a)
q
si
= (kg of sieve)(1.0 kJ/kg K)(T
rg
– T
1
) (6.9b)
q
st
= (lb of steel)(0.12 Btu/lb °F)(T
rg
− T
1
) (6.10a)
q
st
= (kg steel)(0.5 kJ/kg K)(T
rg
– T
1
), (6.10b)
where q
W
, q
si
, and q
st
represent the heat required to desorb the water, heat the
adsorbent, and heat the vessel walls, assuming external insulation, respectively. T
rg
is the regeneration temperature of the gas that enters the bed minus 50°F (28°C),
and T
1
is the inlet gas temperature. The reduction in T
rg
accounts for incomplete
heating of the bed because the regeneration is stopped before the entire bed reaches
the regeneration gas temperature. The total estimated heat of regeneration is then
q = 2.5 (q
W
+ q
si
+ q
st
) (1.10) (6.11)
where the factor of 2.5 corrects for the change in temperature difference (in −
out) across the bed with time during the regeneration cycle; 40% of the heat is
actually transferred to the bed and walls and the rest is lost in the exiting gas.
The 1.10 assumes a 10% heat loss from the system to the surroundings.
Example 6.3 Estimate the heat required to regenerate an adsorption bed that holds
40,000 lb of 4A molecular sieve and 4,400 lb of water. The vessel contains 55,000
lb of steel, and the regeneration temperature is 600°F. The bed operates at 100°F.
Also estimate the gas flow rate under the assumption that the C
P
for the gas is 0.68
Btu/lb-°F, that 60% of the regeneration time involves heating the bed, and that the
bed is on an 8-hour cycle. The gas leaves the regeneration gas heater at 650°F.
Use Equations 6.8 through 6.11 to compute the heat load. The temperature T
rg
=
600 − 50 = 550°F.
Water: q
W
= (1,800 Btu/lb)(lbs of water on the bed) = 1,800 × 4,400 = 7,920 Btu
Sieve: q
si
= (lb of sieve)(0.24 Btu/lb-°F)(T
rg
− T
1
) = 40,000 × 0.24 (550 − 100)
= 4,320 Btu
Steel: q
st
= (lb of steel)(0.12 Btu/lb-°F)(T
rg
− T
1
) = 55,000 × 0.12 (550 −100)
= 2,970 Btu
Use Equation 6.11 to obtain the total heat requirement.
q = 2.5 (q
W
+ q
si
+ q
st
) (1.10) = 2.5 (7,920 + 4,320 + 2,970)(1.10) = 41,800 Btu.
Regeneration Gas Flow Rate Calculation
Time allocated for heating, Θ, is 0.6 × 8 = 4.8 hours. Then the gas flow rate is
m = × q/[C (t – t )] = 41,800/[0.68 4
P Hot B
Θ ..8 (650 – 100)] = 23,300 lb/h.
© 2006 by Taylor and Francis Group, LLC
Gas Dehydration 159
Occasionally the adsorbent beds produce increased fines in the exit gas. The
fines usually result from attrition of the bottom support balls and indicate failure
of the mechanical supports at the bottom of the bed (Richman, 2005).
beds would be run in series, and the second bed would be used as a guard bed.
However, this arrangement is not used that much now, and the primary advantage
of the third bed is that regeneration gas requirements can be reduced by regen-
erating beds placed in series. Also, three beds provide reduced heating require-
ments because regeneration gas can be used to transfer heat from a cooling bed
to a warming bed. Trent (2004) suggests the best approach for four-bed systems
is to run two beds in parallel. For a given gas throughput, the pressure drop is
reduced to one eighth of the two-bed system.
6.3.2.4 Other Factors That Affect the Adsorption Process
Trace amounts of oxygen affect bed life and performance in a variety of ways.
At the normal regeneration gas temperature of 600°F for molecular sieves, 2
moles of oxygen react with methane to form 2 moles of water and 1 mole of
CO
2
. As this reaction is exothermic, higher amounts of oxygen in the gas can
lead to temperatures above the design temperature of the molecular sieve
vessel. When oxygen is present, the temperature of the beds during regener-
ation must be monitored for safety reasons (McCartney, 2005). Clark, et al.
(1975) found that oxygen undergoes partial oxidation reactions with heavier
hydrocarbons, which are adsorbed in the binder, to form alcohols and car-
bocylic acids that ultimately turn to water and CO
2
. If H
2
S is present, it
undergoes oxidation to elemental sulfur, sulfur dioxide, and water. Trent
(2001) points out that oxygen concentrations greater than 20 ppmv also
generate olefins that become coke in the bed. These reactions reduce molecular
sieve capacity by forming solid deposits and by causing incomplete removal
of water during regeneration because the partial pressure of water is higher
To avoid the above reactions as well as COS formation, regeneration tem-
peratures are lowered to the 300 to 375°F range. However, this range increases
the required regeneration time and the amount of regeneration gas used, which
increases recompression cost.
As in all processes, ensuring that the beds are protected from entrained water
and hydrocarbons is important. Even trace amounts of entrained water load the
bed quickly and increase the regeneration heat load. If the gas comes to the
dehydration unit fully saturated, which is often the case, cooling the gas and
removing the condensed water before the gas enters the bed lowers water loading
and potentially increases the drying-cycle time. Trent (2001) notes that cooling
a water-saturated gas by 20°F (10°C) drops the saturated gas-phase water content
by 50%.
The most common bed configuration is the two-bed system shown in Figure
6.8. Trent (2004) notes that many three-bed systems were set up such that two
(see Figure 6.6).
© 2006 by Taylor and Francis Group, LLC
160 Fundamentals of Natural Gas Processing
Example 6.4 Gas, with a molar mass of 23, leaves an amine treater water saturated
at 100°F at 1,000 psia and enters a molecular-sieve dehydrator. How much would
the water load on the adsorption bed be decreased if the gas were cooled to 80°F
before entering the adsorber? Assume that all of the condensed water is removed.
be 64 and 33 lb/MMscf, respectively. The molar mass and temperature are low
enough that the gravity correction is only 1% and can be neglected. The reduction
in water content is (64−33)/64 = 0.48. Thus, if all mist is removed, a 48% reduction
Equation 6.7 that the cooler gas also increases the bed capacity as the temperature-
correction factor increases from 0.94 to 0.99. Thus, cooling the gas by 20°F (1°C)
would essentially double the cycle time.
6.3.3 DESICCANT PROCESSES
In some situations, such as remote gas wells, use of a consumable salt desiccant,
such as CaCl
2
, may be economically feasible. The system can reduce the water
content down to 20 ppmv. Typical salt capacities are 0.3 lb CaCl
2
per lb H
2
O.
The Engineering Data Book (2004b) provides more details.
6.3.4 MEMBRANE PROCESSES
Membranes offer an attractive option for cases in which drying is required to meet
pipeline specifications. Their modular nature, light weight, large turndown ratio,
and low maintenance make them competitive with glycol units in some situations.
Feed pretreatment is a critical component of a membrane process (see
Inlet gas temperature should be at least 20°F (10°C) above the dew point of water
to avoid condensation in the membrane.
Units operate at pressures up to 700 to 1,000 psig (50 – 70 barg) with feed
gases containing 500 to 2,000 ppmv of water. They produce a product gas stream
of 20 to 100 ppmv and 700 to 990 psig (48 to 68 barg). The low-pressure (7 to 60
psig [0.5 to 4 barg]) permeate gas volume is about 3 to 5% of the feed gas volume.
This gas must be recompressed or used in a low-pressure system such as fuel gas.
Smith (2004) suggests that membranes used for natural gas dehydration are
economically viable only when dehydration is combined with acid-gas removal.
On the basis of commercial units installed and several studies (Binci et al. [undated];
Bikin et al., 2003), membranes are economically attractive for dehydration of gas
when flow rates are less than 10 MMscfd (0.3 MMSm
3
/d). Binci et al.(undated)
claim that membrane units are competitive with TEG dehydrators on offshore
platforms at flows below 56 MMscfd (1.6 MMSm
3
/d). Certainly, the reliability and
simplicity of membranes make them attractive for offshore and remote-site appli-
cations, provided the low-pressure permeate gas is used effectively. An added benefit
compared with TEG units is the absence of BTEX emissions with membranes.
Figure 6.1a shows the water content of gas at 1,000 psia and at 100 and 80°F to
of the water load would occur on the adsorption bed. Recall from Figure 6.10 or
Chapter 5). The inlet gas must be free of solids and droplets larger than 3 microns.
© 2006 by Taylor and Francis Group, LLC
Gas Dehydration 161
6.3.5 OTHER PROCESSES
Three relatively new processes are worth mentioning. The first process is a
refrigeration process that mixes methanol with the gas and cools the gas to very
low temperatures. The water−methanol mixture drops out and the methanol is
recovered in a stripper column. The process has several major advantages:
• It can obtain dew points in the −100 to −150°F (−70 to –100°C) range.
• It requires no heat input other than to the methanol regenerator.
• It requires no venting of hydrocarbon-containing vapors.
However, it requires external refrigeration to cool the gas, and minimal
methanol losses occur in the stripper. The Engineering Data Book (2004b) pro-
vides more details.
The second process is the Twister technology, which is discussed in
Yuvancic, 2004) for dehydration because of its simplicity (no moving parts) along
with its small size and weight. Brouwer et al. (2004) discuss the successful
implementation on an offshore platform. Some offshore field pressures are greater
than 2,000 psi (140 bar), so recompression is not needed with the unit where
overall pressure drop is 20 to 30%.
The third process is the vortex tube technology, which also is discussed in
Chapter 7. It also has no moving parts. According to vendor information, it is
used in Europe in conjunction with TEG addition to remove water from gas stored
underground. We found no examples of its use in gas plants.
6.3.6 COMPARISON OF DEHYDRATION PROCESSES
A number of factors should be considered in the evaluation of a dehydration
process or combination of processes. If the gas must be dried for cryogenic liquids
recovery, molecular sieve is the only long-term, proven technology available. It
has the added advantage that it can remove CO
2
at the same time. If CO
2
is being
simultaneously removed, because water displaces CO
2
, the bed must be switched
before the CO
2
breaks through, which is before any water breakthrough. Enhanced
TEG regeneration systems may begin to compete with molecular sieve. Skiff et
al. (2003) claim to have obtained less than 0.1 ppmv water by use of TEG with
a modified regeneration system that uses about 70% of the energy required for
molecular sieves.
High inlet water-vapor concentrations make molecular sieve dehydration
expensive because of the energy consumption in regeneration. Two approaches
are used to reduce the amount of water going to the molecular sieve bed. First,
another dehydration process, (e.g., glycol dehydration) is put in front of the
molecular sieve bed. The second option is to have combined beds with silica gel
or activated alumina in front of the molecular sieve. The bulk of the water is
removed with the first adsorbent, and the molecular sieve removes the remaining
water. This configuration reduces the overall energy required for regeneration.
Chapter 7. It has been considered attractive in offshore applications (Wilson and
© 2006 by Taylor and Francis Group, LLC
162 Fundamentals of Natural Gas Processing
If dehydration is required only to avoid free-water formation or hydrate
formation or to meet the pipeline specification of 4 to 7 lb/MMscf (60 to 110
mg/Sm
3
), any of the above-mentioned processes may be viable. Traditionally,
glycol dehydration has been the process of choice. System constraints dictate
which technology is the best to use. Smith (2004) provides an overview of natural
gas dehydration technology, with an emphasis on glycol dehydration.
When considering susceptibility to inlet feed contamination, one should keep
in mind that replacing a solvent is much easier and cheaper than changing out
an adsorbent bed. However, prevention of contamination by use of properly
designed inlet scrubbers and coalescing filters, if required, is the best solution.
In a conventional gas plant, where inlet fluctuations are handled in inlet receiving,
feed contamination is generally limited to possible carryover from the sweetening
unit. However, in field dehydration the possibility exists of produced water, solids,
oil, and well-treating chemicals entering the dehydrator. Wieninger (1991) dis-
cusses how these components detrimentally affect glycol dehydrators.
6.4 SAFETY AND ENVIRONMENTAL
CONSIDERATIONS
Dehydration processes offer few safety considerations outside of having high-
temperature and high-pressure operations. Probably the most unique safety con-
sideration is when adsorbent beds are being changed. The bed must be thoroughly
purged, preferably with nitrogen, to remove adsorbed hydrocarbons before the
adsorbent is dumped. A potential exists for hydrocarbons on the adsorbent to
ignite when exposed to air because the adsorbent heats as it adsorbs moisture
from the air. Either a highly trained company expert or an adsorbent company
representative should be present to help ensure safe dumping and filling opera-
tions. The dumping process produces dust, and operators must wear protective
clothing and dust masks.
A major environmental concern in dehydration with glycol solutions is BTEX
2
brine must be in accordance with local environmental regulations. Spent adsor-
bents, if properly regenerated, may be discarded into normal land fills. Ethylene
glycol is toxic to humans and must be handled properly.
REFERENCES
Bikin, B., Giglia, S, and Hao, J., Novel Composite Membranes and Process for Natural Gas
Upgrading, Annual Report to Department of Energy, DE- FC26-99FT40497, March
2005.
Binci, F., Ciarapica, F.E., and Giacchetta, G., Natural Gas Dehydration in Offshore Rigs.
Comparison Between Traditional Glycol Plants and Innovative Membrane Sys-
Retrieved July 2005.
emissions. Chapter 9 discusses this issue in more detail. Disposal of spent CaCl
2003, www.osti.gov/bridge/servlets/purl/823967-PH6Fq7/native/, Retrieved July
tems, www.membrane.unsw.edu.au/imstec03/content/papers/IND/imstec033.pdf,
© 2006 by Taylor and Francis Group, LLC
Gas Dehydration 163
Blachman, M. and McHugh, T., Sour Gas Dehydration Technology and Alternatives, Pro-
ceedings of the Laurance Reid Gas Conditioning Conference, Norman, OK, 2000.
Brouwer, J.M., Bakker,G., Verschoof, H.-J., and Epsom, H.D., Supersonic Gas Condition-
ing First Commercial Offshore Experience, Proceedings of the Eighty-Third
Annual Convention of the Gas Processors Association, Tulsa, OK, 2004.
Chi, C.W. and Lee, H., Natural gas purification by 5A molecular sieves and its design
method, AIChE Symposium Series, 69 (134) 95, 1973.
Clark, K.R., Corvini, G., and Bancroft, W.G., Molecular Sieve Treating of Natural Gas
Containing Oxygen, Proceedings of the Laurance Reid Gas Conditioning Confer-
ence, Norman OK, 1975.
Cummings, W. and Chi, C.W., Natural Gas purification with molecular sieves, presented
at the Canadian Chemical Engineering Conf., Calgary, Alberta, Canada, October
23–26, 1977.
de Bruijn, J.N.H. and van Grinsven, P.F.A., Otimizing the on stream times of a mol sieve
dehydration unit, Proceedings of the Laurance Reid Gas Conditioning Conference,
Norman OK, 2001.
Engineering Data Book, 12th ed., Sec. 2, Product Specifications, Gas Processors Supply
Association, Tulsa, OK, 2004a.
Engineering Data Book, 12th ed., Sec. 20, Dehydration, Gas Processors Supply Assocition,
Tulsa, OK, 2004b.
Hernandez-Valencia, V.N., Hlavinka, M.W., and Bullin, J.A., Design Glycol Units for
Maximum Efficiency, Proceedings of the Seventy-First Annual Convention of the
Gas Processors Association, Tulsa, OK, 1992, 310.
Kruka, V.R., private communication, 2005.
Kohl, A. and Nielsen, R., Gas Purification, 5th ed., Gulf Publishing, Houston, TX, 1997.
Lukchis, G.M. Adsorption systems, Part 1: Design by Mass-Transfer Zone concept, Part
II, Equipment Design, and Part III Adsorption Regeneration. UOP Brochure
XF04A.
Malino, H.M., Fundamentals of Adsorptive Dehydration, Proceedings of the Laurance
Reid Gas Conditioning Conference, Norman OK, 2004, 61.
Masaoudi, R., Tohidi, B., Anderson, R., Burgass, R.W., and Yang, J., Experimental measure-
ments and thermodynamic modeling of clathrate hydrate equilibria and salt solubility
in aqueous ethylene glycol and electrolyte solutions, Fluid Phase Equil., 31, 219, 2004.
Parrish, W.R., Won, K.W., and Baltatu, M.E., Phase Behavior of the Triethylene Glycol-
Water System and Dehydration/Regeneration Design for Extremely Low Dew
Point Requirements, Proceedings of the Sixty-Fifth Annual Conversion Gas Pro-
cessors Association, Tulsa, OK, 1986, 202.
Parrish, W.R., Thermodynamic Inhibitors in Brines, Final Report, DeepStar CTR 4210,
Petty, L.E., Practical aspects of molecular sieve unit design and operation. Proceedings of
the Fifty-fifth annual convention of the Gas Processors association, Tulsa OK, 1976.
Prausnitz, J.M., Lichtenthaler, R.N., Azevedo, E.G., Molecular Thermodynamics of Fluid-
Phase Equilibria, Prentice-Hall, Englewood Cliffs, NJ, 1999.
Richman, P., private communication, 2005.
Skiff, T., Szuts, A., Szujo, V., and Toth, A., Drizo Unit Competes with Solid Bed Desiccant
Hydration, Proceedings of the Laurance Reid Gas Conditioning Conference,
Norman OK, 2003, 213.
Dow Chemical Company, Triethylene Glycol, 2003, www.dow.com/PublishedLiterature/
dh_0451/09002f13804518f1.pdf, Retrieved September 2005.
2000. (Contact Adminstrator@DeepStar.org for information.)
© 2006 by Taylor and Francis Group, LLC
164 Fundamentals of Natural Gas Processing
Smith, J.M., Van Ness, H.C., and Abbott, M.M., Introduction to Chemical Engineering
Thermodynamics, 6th ed., McGraw-Hill, New York, 2001.
Smith, R.S., Custom Glycol Units Extend Operating Limits, Proceedings of the Laurance
Reid Gas Conditioning Conference, Norman OK, 1993, 101.
Smith, R.S., Fundamentals of Gas Dehydration Inhibition/Absorption Section, Proceedings
of the Laurance Reid Gas Conditioning Conference, Norman OK, 2004, 17.
Smith, R.S. and Humphrey, S.E., High Purity Glycol Design Parameters And Operating
Experience, Proceedings of the Laurance Reid Gas Conditioning Conference,
Norman OK, 1995, 142.
Trent , R.E., Craig, D.F., and Coleman, R.L., The Practical Application of Special Molec-
ular Sieves to Minimize the Formation of Carbonyl Sulfide During Natural Gas
Dehydration, Proceedings of the Laurance Reid Gas Conditioning Conference,
Norman OK, 1993, 239.
Trent, R.E., Dehydration with Molecular Sieves, Proceedings of the Laurance Reid Gas
Conditioning Conference, Norman OK, 2001.
Trent, R.E., Dehydration with Molecular Sieves, Proceedings of the Laurance Reid Gas
Conditioning Confereence, Norman OK, 2004, 75.
Trimble, H.M., Solubilities of salts in ethylene glycol and in its mixtures with water, Ind.
Eng. Chem. 23, 165, 1931.
Wallace, C.B., Dehydration of supercritical CO2, Proceedings of the Laurance Reid Gas
Conditioning Conference, Norman OK, 1985.
Wichert, E., private communication, 2005.
Wieninger, P., Operating Glycol Dehydration Systems, Proceedings of the Laurance Reid
Gas Conditioning Conference, Norman OK, 1991, 23.
Wilson, J.L. and Yuvancic, J., Process Selection for Dehydrating Gulf of Mexico Offshore
Platform Gas, Proceedings of the Laurance Reid Gas Conditioning Conference,
Norman OK, 2004, 125.
© 2006 by Taylor and Francis Group, LLC
165
7
Hydrocarbon Recovery
7.1 INTRODUCTION
Pipeline quality natural gas specifications include limits on sulfur and water
content, along with higher heating value, which must be about 950 to 1,150
Btu/scf (35,400 to 42,800 kJ/Sm
3
) (Engineering Data Book, 2004a). Exact limits are
set by negotiation between the processor and the purchaser. The previous two chap-
ters addressed the first two specifications. This chapter addresses the heating value.
Field Liquids Removal
Gas & Liquids from Wells
Inlet
Compression
Sulfur
Recovery
Gas Treating
Dehydration
Nitrogen
Rejection
Helium
Recovery
Outlet
Compression
Elemental
Sulfur
Field Dehydration
Field Acid Gas Removal
Field Compression
Sales Gas
Water &
Solids
Natural
Gas Liquids
Natural
Gasoline
Raw Helium
CO
2
Liquefaction
Liquefied
Natural Gas
Inlet Receiving
Hydrocarbon
Recovery
Liquids Processing
© 2006 by Taylor and Francis Group, LLC
166 Fundamentals of Natural Gas Processing
Unless the treated gas contains high concentrations of inerts (N
2
, CO
2
), the heating
value may be too high because of the C
2
+ fraction present. This chapter discusses
hydrocarbon recovery methods to both lower the heating value and create, simul-
taneously, valuable NGL liquid hydrocarbon products.
Processors have additional reasons for reducing the C
2
+ fraction. Hydrocar-
bon recovery frequently is required in field operations for fuel conditioning or
may fuel the compressors with raw gas taken directly from the pipeline. Raw gas
usually is far too rich, and simple systems are used to lower the heating value
(i.e., condition the fuel) by removing heavier hydrocarbons.
Dew point control (or “dew pointing”) is necessary when raw gas lines are
constrained in liquid content as the liquid reduces gas throughput, causes slug-
ging, and interferes with gas metering. This situation can be a problem for offshore
production, where several operators share a common gas line to shore. These
lines may operate at 1,800 psig (125 barg) or higher and be at seafloor temper-
atures of 36 to 40°F (2 to 4°C). Dew point control is also necessary if a potential
for condensation is present in a process because of temperature or pressure drops.
The latter happens when the gas is in the retrograde condensation region. How-
ever, effective dew point control is much less demanding than C
2
+ recovery, as
it can be accomplished without removal of a large portion of the C
3
+ fraction.
After a brief discussion of retrograde condensation, this chapter discusses some
common process elements used in conventional hydrocarbon recovery and then
describes some common processes that utilize these elements. A myriad of process
figurations. The chapter also includes brief descriptions of some new technologies.
7.1.1 RETROGRADE CONDENSATION
A major reason for dew point control is the fact that rich natural gas mixtures that
contain heavier hydrocarbons exhibit a nonintuitive behavior called retrograde con-
(but typical) natural gas mixture that contains 85 mol% methane with 4.8 mol% of
C
3
+. The envelope is the bubble point–dew point line of the mixture. At any tem-
perature and pressure combination outside the envelope, the mixture is single phase.
At temperatures and pressures inside the envelope, two phases exist.
Three points on the envelope are important:
• The cricondentherm, the maximum temperature at which two phases
can exist
• The cricondenbar, the maximum pressure at which two phases can exist
• The critical point, the temperature and pressure where the liquid and
vapor phases have the same concentration
The retrograde condensation effect can be seen by following the vertical dashed
line in Figure 7.1. For the mixture at the temperature and pressure at the top of the
densation. Figure 7.1 shows the pressure−temperature relationship for a hypothetical
dew point control. As noted in Chapter 3, booster stations in gathering systems
variations are available, and this chapter illustrates only a few simple process con-
© 2006 by Taylor and Francis Group, LLC
Hydrocarbon Recovery 167
line, a single phase exists. Dropping the pressure causes a liquid phase to form
(retrograde condensation), which will be present until the pressure is below the
envelope. The dotted-line path is similar to what happens in a pipeline because
of line pressure drop if pipeline temperature is constant.
The dashed curve inside the envelope denotes the pressure and temperature
of the mixture when the vapor quality is 95 mol%. This curve shows the dramatic
effect on the phase behavior from condensation of only 5 mol% of the vapor.
The cricondentherm of this vapor phase is about 50°F (30°C) lower than the
original mixture, and condensation at typical pipeline temperatures would not be
possible.
As can be seen from the previous paragraph, the cricondentherm of a mixture
strongly depends on the molecular weight of the heavy components. The cricon-
denbar increases with increased molecular weight. Concentration of the heavy
components present is relatively less important than their molecular weight. On
the basis of simple flash calculations, a mixture that contains methane and 10
mol% propane has a cricondentherm comparable to a methane–heptane mixture
with only 0.06 mol% heptane.
7.2 PROCESS COMPONENTS
The process elements involved in hydrocarbon recovery vary, depending upon
the desired products and gas volume being processed as well as inlet composition
and pressure. Not all of the elements listed here will be found in all plants, and
often it is a choice of one component or another.
FIGURE 7.1 Pressure−temperature diagram for a hypothetical raw natural gas that con-
tains predominately methane, with trace components up to heptane. The dashed curve
represents the vapor-phase line at 95% quality. Points A, B, and C denote the criconden-
therm, cricondenbar, and critical point of the mixture, respectively.
0
250
500
750
1000
1250
1500
−300 −250 −200 −150 −100 −50 0 50 100
Temperature, °F
P
r
e
s
s
u
r
e
,

p
s
i
a
C
B
A
Liquid
Vapor
© 2006 by Taylor and Francis Group, LLC
168 Fundamentals of Natural Gas Processing
7.2.1 EXTERNAL REFRIGERATION
External refrigeration plays a major role in many hydrocarbon recovery processes,
as it is used to cool the gas stream to recover a significant amount of C
3
+ and to lower
gas temperatures as the gas goes into other stages of hydrocarbon recovery. It may
be the only source of refrigeration when inlet pressure is low. Adsorption and vapor-
compression refrigeration are used in special situations. However, vapor compression
using propane as the working fluid is the most common in gas plants and will be
discussed here. (LNG facilities also use ethane or ethylene as well as hydrocarbon
discussed first and is then followed by a look at process variations. The Engineering
Data Book (2004d) provides valuable design guidelines that are omitted here.
7.2.1.1 Basic Propane Refrigeration Process
As described in any basic thermodynamics book (e.g., Smith et al., 2001), the
refrigeration cycle consists of four steps that are depicted on the pressure−
enthalpy chart in Figure 7.2:
1. Compression of saturated refrigerant vapor at point A to a pressure
well above its vapor pressure at ambient temperature at point B
2. Condensation to point C by heat exchange with a cooling fluid, usually air
3. Expansion through a valve (Joule-Thomson expansion) to cool and
condense the refrigerant to point D
4. Heat exchange with the fluid to be cooled by evaporation of the refrig-
erant back to point A
FIGURE 7.2 Schematic of refrigeration cycle on a pressure−enthalpy chart.
B C
D
Enthalpy
P
r
e
s
s
u
r
e
Critical point
Vapor
phase
Liquid
phase
B′
A
mixtures as refrigerants. See Chapter 13 for details.) The basic refrigeration cycle is
© 2006 by Taylor and Francis Group, LLC
Hydrocarbon Recovery 169
Figure 7.3 shows the flow diagram for a single-stage propane refrigeration
system, with typical operating conditions. Each of the steps is described below.
Compression Step—Cycle analysis begins with propane vapor entering
the compressor as a vapor at 14.5 psia (1 bar) and approximately −40°F (−40°C),
The power required and compressor discharge temperature depends upon
reciprocating compressors, but oil-injected screw compressors are now pre-
ferred because they can complete the compression in one stage. (Large refrig-
eration units such as those used in LNG plants use centrifugal compressors.)
The work of compression is simply
w
S
= (h
B
– h
A
)/h
IS
, (7.1)
where h
IS
is the adiabatic efficiency of the

compressor (see Chapter 4). Taking
into account compressor nonideality, the actual enthalpy at the end of the expansion is
h
B′
= h
A
+ w
S
. (7.2)
Compressor power to the refrigeration system is the product of the mass flow
rate and shaft work
. (7.3)
FIGURE 7.3 Single-stage propane refrigeration system. (Adapted from Engineering Data
Book, 2004d.)
Condenser
Chiller/evaporater
−40°F, 16 psia
J-T valve
Receiver
~120°F, 240 psia
Compressor
Suction
drum
250 psia
HP m w
S
=

where it is compressed to 250 psia (17 bar) (point A to point B in Figure 7.2).
compressor efficiency (see Chapter 4 for details). Plants once used multistaged
© 2006 by Taylor and Francis Group, LLC
170 Fundamentals of Natural Gas Processing
Condensation Step—The warm gas goes to an air- or water-cooled con-
denser, where the propane cools to 100 to 120°F (38 to 50°C), totally condenses,
q
Cond
= h
C
– h
B′
. (7.4)
Expansion Step—Propane liquid leaves the receiver and flashes through a
J-T valve, where the temperature and pressure drop to −40°F (−40°C) and 16 psia
(1 bar) (point C to point D). No change occurs in the enthalpy, but the temperature
drops to the saturation temperature of the liquid at the expansion-discharge pressure,
and h
C
= h
D
if there are no heat leaks. If there is a heat leak, , then
h
D
= h
C
+ . (7.5)
The fraction, f, of propane condensed is computed knowing the initial enthalpy
and liquid and vapor enthalpies at the condensation temperature, which for the
given case is
.
Assuming the vapor leaves the chiller as a saturated vapor,
= h
A

. (7.6)
Refrigeration Step—The cold propane then goes to a heat exchanger, the chiller,
where it cools the process stream by evaporation (point D to point A in Figure 7.2).
Because the propane in the chiller is evaporating, and a minimal heat exchange
occurs between cold propane vapor and the inlet gas, the inlet and outlet propane
temperature remains constant. The propane returns to the compressor suction slightly
above −40ºF (−40°C). The heat absorbed by the propane is simply h
A
− h
D
.
In the past, most chillers were the kettle type in which propane is on the shell
side and the liquid level is maintained above the tube bundle. Now, other high-
details). The chiller typically has two zones of heat transfer. The first is exchange
of boiling propane with gas above its dew point and will involve only sensible
heat. The second zone has condensing vapors from the process stream and boiling
propane, which gives a much higher overall heat-transfer coefficient (Engineering
Data Book, 2004d).
To complete the cycle, the propane vapors leave the chiller and go to the
suction drum before being compressed again.
q
L
q
L
h h f f h
D D
L
D
V
= + − ( ) 1
h
D
V
f h h h h
A D A D
L
= − − ( ) ( )
and collects in a receiver (point B′ to point C in Figure 7.2). This step is simply
performance heat exchangers (e.g., plate-fin) are used (see Section 7.2.3 for more
© 2006 by Taylor and Francis Group, LLC
Hydrocarbon Recovery 171
Refrigeration-cycle performance is commonly stated in terms of coefficient
of performance (COP), which is the ratio of the refrigeration obtained divided
. (7.7)
Example 7.1 Compute the liquid fraction produced and the COP for the propane
leak into the system and assume the compressor efficiency is 77%. Use the saturation
Compression Step—Following the cycle as given above, first calculate the work
of compression. To compute compressor work, we use Equation 3.1 and assume
that the inlet-gas condition is saturated vapor at –40°F. The work per unit mass
(assuming reversible compression) required to compress the propane from 14.5 to
w
S
= (h
B
– h
A
)/h
IS
= (h
250 psia
– h
14.5 psia, −40°F
)/h
IS
=(240 – 182)/0.77 = 75 Btu/lb.
Because of the compressor inefficiency, the enthalpy of the vapor that leaves the
compressor is
h
B′
= h
250 psia
= h
14.5psia,−40°F
+

w
S
= 182 + 75 = 257 Btu/lb,
which corresponds to about 140°F at 250 psia.
Condensation Step—Pressure drop between the condenser inlet and receiver is
usually around 8 to 10 psi (0.55 to 0.7 bar). The heat load on the condenser is the
change in enthalpy from the heated vapor to condense to all liquid. On the basis of
the saturation table, this value is
q = h
C
– h
B
′ = h
250 psia
– h
242psia, 120°F
= 257 – 98.7 = 158 Btu/lb.
To calculate the mass fraction condensed use Equation 7.6
=(182.2 – 98.7)/(182.2 – 0) = 0.46
The COP is, from Equation 7.7:

= 0.46 (182.2 – 0)/75 = 1.1
COP h h h h f h h w
A D B A A D
L
S
= − − = − ( ) ( ) ( )
f h h h h h h
A D A D
L
F
V
psia F
= − − = −
− ° °
( ) ( ) ( ) (
, 40 240 120
hh h
F
V
F
L
− ° − °

40 40
)
COP f h h w f h h w
A D
L
S F
V
F
L
S
= − = −
− ° − °
( ) ( )
40 40
by the work required. On the basis of Figure 7.2, the COP is determined by
refrigeration system on the basis of the conditions given in Figure 7.3. Ignore heat
table and PH diagram for propane given in Appendix B for the calculations.
250 psia (from Appendix B) is
© 2006 by Taylor and Francis Group, LLC
172 Fundamentals of Natural Gas Processing
7.2.1.2 Alternate Process Configurations
Thermodynamics dictate that to minimize the refrigeration work (i.e., compression
required) heat from the chiller should be removed at as high a temperature as
possible. One way to reduce compressor duty per unit of refrigeration duty is to
multistage the refrigeration process by removal of process heat at more than one
temperature. Figure 7.4 shows a two-stage system, with representative operating
conditions. In this system, the condensed propane stream expands to about 62 psia
(4.3 bar) and 25°F (−4°C) in two parallel J-T valves. One of the expanded streams
goes to a chiller before going to the suction surge drum for the second stage of
the compressor.
Vapor from the surge drum goes to the second stage of the compressor, while
the liquid goes through a second J-T expansion to provide the low-temperature
cooling. Pressures are adjusted so that the compression ratio is equal in both
but less of a benefit when going to three stages.
Multistaging reduces work requirements by removing heat from the process
stream at different temperatures. An alternative is removal of heat from the
refrigerant before it is expanded. Refrigerant subcooling is sometimes used by
exchange of the propane that leaves the receiver with a portion of the cold liquid
propane. An Engineering Data Book (2004d) example shows that a 10% reduction
in recirculation rate to the second stage of two-stage refrigeration is obtainable
when the propane from the condenser is cooled 10°F (5°C).
FIGURE 7.4 Two-stage propane refrigeration system, with second heat exchanger and
economizer. Units may omit either the first stage heat exchanger or expansion directly to
the economizer.
~120°F
240 psia
−40°F
16 psia
60 psia 250 psia
Condenser
J-T valve
J-T Valve J-T valve
Receiver
25°F
62 psia
Interstage
Economizer
Chiller/
Evaporator
25°F, 62 psia
Chiller/Evaporator
−40°F, 16 psia
compressor stages. Table 7.1 shows the significant savings in going to two stages
© 2006 by Taylor and Francis Group, LLC
Hydrocarbon Recovery 173
Although a second heat exchanger provides the most benefit from staging a
refrigeration system, savings can be obtained by use of only a single J-T expansion
into the economizer or by use of single J-T expansion and the heat exchanger. Gas
plants commonly omit the first-stage heat exchanger for cost purposes. However,
when energy requirements are critical, such as in an LNG plant, the J-T valve that
bypasses the heat exchanger is commonly eliminated and three stages of refrigeration
Another way to alter the temperature at which heat is removed is to use
refrigerant cascading. In this option, one refrigerant (e.g., propane) is used to
remove heat from another refrigerant (e.g., ethane), which then cools the process
gas down to as low as −120°F (−85°C). This technique is commonly used in
LNG plants and is discussed in Chapter 13. Many gas plant process stream
temperatures go well below −120°F (−85°C) but usually rely upon process gas
expansion for cooling instead of additional external cooling.
Obviously many combinations are possible for making refrigeration more
efficient. However, each must be balanced with the associated additional capital
cost, realizable operating cost savings, and operating complexity. Although little
incentive exists for more complex systems in gas processing, LNG plants make
extensive use of complex refrigeration cycles, as discussed in Chapter 13, to
reduce refrigeration (i.e., compression) costs.
7.2.1.3 Effect of Operating Variables on Refrigeration
Performance
A number of other process variables affect refrigeration capacity. An obvious one is
the condenser outlet temperature, which is limited by the heat removed in the
condenser and dictates compressor discharge pressure. If air cooling is used, the
temperature there can exhibit wide swings, both seasonal and daily, in ambient
perature plays on both condenser and compressor duties to maintain a constant
TABLE 7.1
Effect of Multistaging on Condenser and Compression Duty
for Constant Refrigeration Duty with Propane as the Refrigerant
Number of Stages
1 2 3
Change in compressor power (%) 0 −19.2 −23.3
Change in condenser duty (%) 0 −8.2 −9.6
Refrigeration temperature is −40°F (−40°C) and condensing temperature is 100°F
(38°C).
Source: Engineering Data Book (2004d).
and condenser temperatures. Table 7.2 shows the importance the condensing tem-
are common. See Chapter 13 for details of more complex heat exchanger systems.
© 2006 by Taylor and Francis Group, LLC
174 Fundamentals of Natural Gas Processing
refrigeration capacity when the refrigerant is propylene; results should be com-
parable for propane.
Another factor that affects performance is refrigerant purity. Propane com-
monly contains small amounts of ethane. Low concentrations pose no problem.
Ethane accumulates in the propane receiver and increases both condensation pres-
sure and temperature. If refrigeration capacity is adversely affected, the propane
can be purged into the plant-gas inlet stream for recovery and the system recharged.
Whereas light ends affect performance, heavy components are less of a
problem but still must be removed. The most common heavy material is com-
pressor oil. Heavy liquids are removed by periodically taking a slip stream from
the bottom of the chiller, which goes to a small heated vessel where the propane
is evaporated and returned to the suction of the compressor. The heat source is
hot propane vapor from compressor discharge. Butanes may also be present, but
unless in large quantities they will not significantly raise the chiller temperature.
7.2.2 TURBOEXPANSION
Until the 1960s, Joule-Thomson expansion was the only way used to cool gas-
plant streams by pressure drop. Herrin (1966) describes the first turboexpander
plant. The J-T valve, which is essentially a control valve with a variable or fixed
orifice, is an extremely simple, inexpensive, and widely used means to reduce
gas temperature. Although still extensively used in many applications to produce
refrigeration, J-Ts are being widely supplanted by turboexpanders in gas plants
for cooling the process stream when it is a gas. Turboexpanders are, in essence,
centrifugal compressors that run backwards. Unlike J-T expanders, they perform
work during the process. Whereas J-T expansion is essentially an isenthalpic
process (therefore, no work is done on or by the gas), an ideal, thermodynamically
reversible turboexpander is isentropic. The maximum reversible work required
for compression is isentropic, and, conversely, the maximum reversible work
TABLE 7.2
Effect of Condensed Refrigerant Temperature on Condenser and
Compression Duty for Constant Refrigeration Duty with Propylene
As the Refrigerant
Condensing
Temperature, °F (°C)
60 (15) 80 (27) 100 (38) 120 (49) 140 (60)
Change in
Compressor duty (%)
−36.6 −19.8 Base 28.8 66.4
Change in
Condenser duty (%)
−16.3 −8.7 Base 13.6 31.5
Source: Engineering Data Book (2004d).
© 2006 by Taylor and Francis Group, LLC
Hydrocarbon Recovery 175
recovered by a turboexpander system on expansion is also isentropic. Turboex-
pansion provides the maximum amount of heat removal from a system for a given
pressure drop while generating useful work. The work is used to drive compres-
apply to the turboexpander.
The major breakthrough for turboexpanders came when the design and mate-
rials made it possible for condensation to occur inside the expander. The fraction
condensed can be up to 50% by weight (Jumonville, 2004). However, the droplets
must generally be 20 microns in diameter, or less, as larger droplets cause rapid
erosion of internal components.
Most turboexpanders drive centrifugal compressors to provide a portion of
the outlet compression. In situations where inlet pressures are very high (e.g.,
offshore) turboexpanders are used in pressure letdown to provide refrigeration
for dew point control and to generate power.
Like compressors, expanders can be positive displacement or dynamic;
dynamic can be radial or axial. Reciprocating expanders were used for liquefying
radial unit with inward (centripetal) flow, and discussion is restricted to this type.
A cutaway view of a typical turboexpander for gas processing is shown in
Figure 7.5. The expander is the unit on the right, and the compressor is the unit
on the left. Gas enters the expander through the pipe at the top right, and is guided
onto the wheel by the aerodynamically shaped adjustable guide vanes, which
completely surround the expander wheel. The swirling high-velocity inlet gas
turns the wheel and transfers part of its kinetic energy to the wheel and shaft,
and exits to the right through the tapered nozzle. Because part of the energy of
the gas has been transferred to the wheel, the exit gas is at a much lower
temperature and pressure than the gas entering. The expander wheel, directly
coupled to the compressor wheel, provides the work necessary to drive the
centrifugal compressor on the left. Low-pressure gas enters in the straight section
FIGURE 7.5 Cutaway view of a turboexpander. (Courtesy of Mafi-Trench Corp.)
sors or electrical generators. The equations in Chapter 4 for isentropic compression
gases (see Chapter 13). However, the only type used in gas processing is the
© 2006 by Taylor and Francis Group, LLC
176 Fundamentals of Natural Gas Processing
on the left, is compressed by the compressor wheel, and exits at the top of the
unit. Lubricating oil enters in the top port shown in the center of the unit.
About 50% of the enthalpy change occurs in the turbine (Jumonville, 2004).
The increase in velocity over the vanes results in the other 50% of the total
pressure and temperature drop across the expander unit. Thus, the inlet guide
vanes are a vital part of the energy conversion process in a turboexpander. The
high velocity of the gas that exits the vanes and enters the expander wheel greatly
improves overall turboexpander efficiency (Peranteaux, 2005).
conditions for turboexpanders vary, depending on the process application and the
general idea of turboexpander operations in a gas plant. Note the size of the wheel
in Table 7.3 and the rotating speeds in both tables.
Although simple in principle, turboexpanders are sophisticated, yet
extremely reliable pieces of rotating machinery. In addition to the complex
aerodynamic design of the expander and compressor components of the system,
the unit must be equipped with low drag but 100% effective shaft seals. These
seals operate under high pressure and at temperatures that range from cryogenic
to well above ambient. By injecting clean, dry “seal gas” into the shaft seals,
lube oil is prevented from escaping the bearing housing, and the bearings are
protected from the cold process gas. Fluid-film bearings that support the shaft
that connects the expander and compressor sections traditionally are supplied
with oil by use of a pressurized recirculating system. The Engineering Data
Book (2004c) provides more details.
FIGURE 7.6 Turboexpander in large gas plant. The turboexpander processes 30,500
lb/min (830,000 kg/h) and recovers approximately 9.2 MW (12,000 hp) of power. (Cour-
tesy of Mafi-Trench Corp.)
composition of the gas being processed. However, Tables 7.3 and 7.4 provide a
Figure 7.6 shows a large turboexpander used in a large gas plant. Operating
© 2006 by Taylor and Francis Group, LLC
Hydrocarbon Recovery 177
In principle, both the turboexpander and compressor can be multistaged.
However, to date, mechanical sealing problems have made multistaging imprac-
tical for both the turbine and expander (Peranteaux, 2005).
Many new units employ magnetic bearings instead of fluid-film bearings. Mag-
netic bearings keep the shaft in place by constantly adjusting the current in radial
and axial electromagnets that support the shaft. Although the initial investment in
magnetic bearings is slightly higher than for comparably sized fluid-film bear-
ings, this design provides substantial savings in lifetime utility operating costs
TABLE 7.3
Expander−Compressor Design-Point Conditions
Expander Compressor
Inlet gas rate, MMscfd (MMSm
3
/d) 250 (7.1) 187 (5.3)
Molar mass 22.80 19.69
Inlet pressure, psia (bar) 1,590 (109.7) 855 (58.9)
Inlet temperature, °F (°C) 30 (−1.0) 34 (1.0)
Outlet pressure, psia (bar) 870 (60.1) 1,165 (80.3)
Outlet temperature, °F (°C) −15 (−26.1) 78 (25.4)
Liquid formation, wt% 36 —
Power, kW 2,004 1,986
As tested efficiency, % 85.0 78.5
Speed, rpm 16,700 16,700
Wheel diameter, in. (mm) 7.75 (197) 10.9 (277)
Source: Lillard and Nicoll (1994).
TABLE 7.4
Turboexpander Normal Design Conditions
Expander Compressor
Inlet flow rate, lb/h (kg/h) 221,000 (100,243) 208,000 (94,347)
Inlet gas rate, MMscfd (MMSm
3
/d) 115.6 (3.27) 115.2 (3.26)
Molar mass 17.46 16.5
Inlet pressure, psia (bar) 1,080 (74.5) 470 (32.4)
Inlet temperature, °F (°C) −50 (−46) 60 (16)
Outlet pressure, psia (bar) 480 (33.1) 576 (39.7)
Outlet temperature,°F (°C) −113.6 (−80.9) 93.5 (34.2)
Liquid formation, wt% 19.3 —
Efficiency, % 83 74
Speed, rpm 22,000 22,000
Power, bhp (bkW) 1,380 (1,029) 1,350 (1,007)
Source: Agahi and Ershaghi (1992).
© 2006 by Taylor and Francis Group, LLC
178 Fundamentals of Natural Gas Processing
when compared with conventional oil-lubricating systems. More importantly, they
eliminate the potential for oil contamination in the cold section of the plant
(Jumonville, 2004). Nonetheless, magnetic bearing systems still require seal gas.
In this case, the seal gas protects the bearings from the cold process gas while
simultaneously ensuring adequate cooling for the bearing coils (Peranteaux, 2005).
The Engineering Data Book (2004c) emphasizes some points that should be
kept in mind for turboexpanders:
• Entrainment. Gas that enters the turboexpander must be free of both
solids and liquids. Fine-mesh screens are used to protect the device,
and the pressure drop across the screen should be monitored.
• Seal gas. This gas isolates process gas from the lubricating oil, or
isolates process gas from the shaft if magnetic bearings are used, and
must be clean and constantly available at the operating pressure. Sales
gas is commonly used. Otherwise, a warmed inlet gas stream off of
the expander inlet separator is used. (The gas must be warmed to 70°F
[20°C] or more to prevent thickening of the lube oil, if used.)
• Lubricant pumps. These pumps must maintain a constant flow to lubri-
cate the bearings if oil is used. A spare pump is mandatory. The
Engineering Data Book (2004c) describes the lubrication system.
• Shut-off valves. A quick-closure shut-off valve is used to shut in the
inlet for startup and shutdown.
As is the case for centrifugal compressors, turboexpander efficiency diminishes
when operating off of the design point. This variance can be about 5 to 7 percentage
points when the flow increases or decreases by 50%. However, the turboexpander
normally is driving a compressor, which also will suffer loss in efficiency when off
of the design point. Therefore, the overall effect on the turboexpander−compressor
unit efficiency will be larger. As with centrifugal compressors, surge control is needed.
compute the outlet temperature and work generated per lb of gas inlet. For this
do the calculations. Compare this result to the outlet temperature and work done if
a J-T expansion occurs over the same pressure drop and inlet temperature.
Turboexpander Calculations Inlet conditions of the gas are −50°F and 1,080
psia. At these conditions, for pure methane, the enthalpy h = 255 Btu/lb, and the
entropy s = 0.69 Btu/lb-°R. The ideal expansion is isentropic, so we follow the
constant entropy line to the exit pressure of 480 psia, which is close to the
saturation boundary of methane, where the temperature is −138°F and h = 230
Btu/lb. The work done is (255 – 230) = 25 Btu/lb. However, the expander is only
77% efficient, so the actual work done is 0.77 (25) = 19.2 Btu/lb. The corrected
final enthalpy is then 255 – 19.2 = 236 Btu/lb at 480 psia, which corresponds to
about −135°F. Therefore, the inefficiency did not greatly affect the final temperature
Example 7.2 Use the inlet conditions and outlet pressure supplied in Table 7.4 to
example, assume the gas is pure methane and use diagrams from Appendix B to
© 2006 by Taylor and Francis Group, LLC
Hydrocarbon Recovery 179
in this example. The effect of the efficiency in plants that process large quantities
of gas is significant, both in cooling and in lost power for recompression.
J-T Expansion The J-T expansion follows the constant enthalpy line down to the
exit pressure of 480 psia, giving an exit temperature of –115°F. The work obtained
during expansion is zero.
Therefore, use of a turboexpander with 77% efficiency provides an outlet gas that is
20°F colder and at the same time provides 19.2 Btu/lb of work. The benefit of
turboexpansion compared with J-T expansion is noticeable in plants when the tur-
boexpander is bypassed and J-T expansion is used. In such cases, the NGL production
can drop by as much as 80%. Although this comparison is slightly unfair because the
plant design is optimized for turboexpansion, it points out the value of turboexpanders.
7.2.3 HEAT EXCHANGE
Most heat exchangers in a gas plant operating at or above ambient temperature
are conventional shell and tube type and are ideal for steam and hot oil systems
where fouling occurs. They are relatively inexpensive and easy to maintain
because the tube bundle can be removed and tubes cleaned or replaced as needed.
The Engineering Data Book (2004b) provides extensive details of the exchangers
for use in gas processing. Where the fluids are clean and fouling does not occur,
such as in gas−gas exchangers, compact heat exchangers are ideal. Wadekar
(2000) provides a good overview of the more common types. This section briefly
discusses two kinds, brazed-aluminum plate-fin heat exchangers and printed
circuit heat exchangers, which commonly are used in gas processing. Because
the literature for conventional shell and tube heat exchangers is extensive. these
exchangers are not discussed.
7.2.3.1 Plate-Fin Exchangers
Cryogenic facilities have made extensive use of brazed-aluminum plate-fin heat
exchangers since the 1950s. Instead of a shell and tube configuration, these units
consist of channels formed by a thin sheet of aluminum pressed into a corrugated
pattern (the fin) sandwiched between two aluminum plates. Each layer resembles
the end view of corrugated cardboard. The fin channels may be straight or may
have a ruffled or louvered pattern to interrupt the straight flow path.
Advantages of plate-fin exchangers include (Engineering Data Book, 2004b):
• Light weight.
• Excellent mechanical strength at subambient temperatures (used in
liquid helium service [−452ºF (−268ºC)]). Can operate at pressures up
to 1,400 psig (96 barg).
• High heat transfer surface area. Up to six times the surface area per unit
volume of shell and tube exchanger and 25 times the area per unit mass.
© 2006 by Taylor and Francis Group, LLC
180 Fundamentals of Natural Gas Processing
• Complex flow configurations. Can handle more then 10 inlet streams
with countercurrent, crossflow, and counter crossflow configurations.
• Close temperature approaches. Temperatures of 3°F (1.7°C) for single-
phase fluids compared with 10 to 15ºF (6 to 9ºC) for shell and tube
exchangers and 5°F (2.8°C) for two-phase systems.
Drawbacks and limitations of the exchangers include:
• Single-unit construction. Repair can be more costly and time consum-
ing than with shell and tube exchangers.
• Maximum operating temperature of approximately 150°F (~85°C),
although special designs go to 400°F (205°C).
• Narrow channels. More susceptible to plugging, and fine mesh screens
are needed where solids may enter. Components that might freeze out,
water, CO
2
, benzene, and p-xylene, must be in sufficiently low con-
centrations to avoid plugging. The exchangers can be difficult to clean
if plugging occurs.
• Less rugged. Does not accept rough handling or high pipe stress on nozzles.
• Limited to fluids noncorrosive to aluminum. Caustic chemicals are
corrosive but not corroded by acid gases, unless free water is present.
• Susceptible to mercury contamination. Mercury amalgamates with alu-
minum to destroy mechanical strength.
• Susceptible to thermal shock. Maximum rate of temperature change is
4°F/min (2°C/min ), and maximum difference between two streams is
55°F (30°C) (Howard, 1998).
The Engineering Data Book (2004b) provides some additional information
on plate-fin exchangers, including fin geometries. Lunsford (1996) provides
some guidelines on designing plate-fin exchangers. However, because of the
complex design, heat exchanger vendors should be contacted for details on a
specific application. Most often, heat transfer surface areas will be given for
both hot and cold sides. This practice is comparable to adding both the shell
and tube heat transfer surface areas together in a conventional shell and tube
exchanger.
7.2.3.2 Printed Circuit Heat Exchangers
Another heat exchanger type, the printed circuit heat exchanger (PCHE) is used
in clean service (Engineering Data Book, 2004b). This technology is relatively
new, commercialized in the 1980s, but hundreds of units are in service (Pua and
Rumbold, 2003). Like electronic printed circuits, heat transfer passages are etched
in plates, and the plates are bonded together by diffusion bonding. Unlike the
brazed-aluminum exchangers, they are rugged and, depending on materials of
construction, go to high temperatures and pressures but can still handle complex
flow schemes that involve many streams. Heat transfer passage sizes range from
“microchannels” (less than 8 mil, 200 microns) to “minichannels” (0.12 in, 3 mm)
© 2006 by Taylor and Francis Group, LLC
Hydrocarbon Recovery 181
to provide high heat transfer surface areas. Heat transfer area per unit volume
can be 800 compared with 500 for plate-fin exchangers. Like plate-fin exchangers,
vendor design is required.
Offshore operations employ PCHEs in many applications because they offer
comparable heat transfer at comparable pressure drops at significantly less size
and at one fifth the weight (Pua and Rumbold, 2003). One example of a high-
temperature onshore application is in gas turbine-driven compressor stations
where hot lube oil is used to preheat fuel gas to the gas turbine to avoid conden-
sation problems (Sell et al., 2004). Haynes and Johnson (2002) discuss exchanger
performance characteristics.
7.2.4 FRACTIONATION
In addition to conventional distillation columns, two other types of distillation
columns are commonly found in gas plants: stabilizers and demethanizers. Stabi-
lizers are stripping columns used to remove light ends from NGL streams. Demeth-
anizers are also stripping columns to remove methane from the NGL bottoms
product. Demethanizers also act as the final cold separator, a collector of cold NGL
liquids, and source of recovering some refrigeration by cooling warm inlet streams.
7.2.4.1 Stabilizers
The primary focus of dew pointing or fuel conditioning is to obtain a leaner gas.
However, the “by-product” is a liquid phase that contains a substantial amount
of volatiles. To make the liquid product easier to store and to recover more light
ends for fuel or sales gas, many of the systems will “stabilize” the liquid by
passing it through a stabilizer column. The stabilizer feed typically enters at the
top of a packed or tray column and no reflux occurs. To increase stripping of
light ends, the column pressure will be lower than that of the gas separator that
feeds the column. In some cases, a stripping gas may be added near the bottom
of the column in addition to the externally heated reboiler installed to provide
additional vapor flow and enhance light-ends removal. This feature usually comes
as an increased operating cost because the gas from the stripper is at low pressure
and must be recompressed if put back into the inlet gas stream upstream of the
gas treating unit.
7.2.4.2 Demethanizer
A distinguishing feature of gas plants with high ethane-recovery rates is the
columns in the following ways:
• It has an increased diameter at the top to accommodate the predomi-
nately vapor feed to the top tray.
• It is typically primarily a stripping column, with no traditional
condenser–reflux stream.
demethanizer, as shown in Figure 7.7. The column differs from usual distillation
© 2006 by Taylor and Francis Group, LLC
182 Fundamentals of Natural Gas Processing
• It may have several liquid feed inlets further down the column that
come from low-temperature separators.
• It may have several side reboilers, the primary purpose of which is to
cool the gas going through the reboiler to recover some of the refrig-
eration available in the warming NGL stream.
• It has a large temperature gradient; over 170°F (75°C) is common.
The column serves two main functions: it acts as a flash drum for the top
feed, which comes in as a cold, two-phase stream, and it removes methane
from the bottoms product. Depending upon the plant configuration, the feed
may be from a turboexpander, a J-T valve, or a heat exchanger. In some
configurations the columns have reflux, but many demethanizers have no reflux.
The NGL bottoms product is usually continuously monitored for methane
content, which typically is kept below 0.5 liquid vol% of the ethane, on a C
3
+
free basis.
The top of the column usually operates in the −175 to −165°F (−115 to −110°C)
range, with pressures in the 200 to 400 psig (14 to 28 barg) range. Operation of
FIGURE 7.7 Demethanizer column.
© 2006 by Taylor and Francis Group, LLC
Hydrocarbon Recovery 183
the demethanizer at higher pressures reduces compression costs. Shah and Stucky
(2004) discuss the advantages and disadvantages of higher pressure operation.
To obtain high ethane recovery, the column must operate near the low end
of the temperature range, which translates into a higher compression load. The
demethanizer usually has 18 to 26 trays, which operate at a tray efficiency of 45
to 60% (Engineering Data Book, 2004f) or an equivalent packed bed.
7.3 RECOVERY PROCESSES
Many process configurations are used to recover hydrocarbons in the field and
in gas plants. The best configuration depends upon many variables, including:
• Product slate
• Gas volumes
• Gas composition
• Pressures, both inlet and outlet
The product slate dictates the required lowest operating temperature of the
gas. Both dew point control and fuel conditioning have the same main product—
a residue gas with reduced C
3
+ fraction. Dew point control is usually a field
operation, and stabilization of the produced liquid is site specific. Although gas
temperatures in a low-temperature separator (LTS) may go down to −40°F (−40°C),
only a cold separator is required to separate the light ends from the liquid. Two
new technologies, Twister and vortex tube, discussed below make dew point
control and fuel conditioning a one-step process. In addition, use of membranes
for fuel conditioning is discussed briefly.
If limited ethane recovery (<60% ethane) is desired, the recovery process
is essentially a low temperature separator, except that fractionation of the cold
liquid is added to increase the recovery. Lean oil absorption is sometimes used
for up to approximately 50% ethane recovery. For high ethane recovery, the
gas processing temperatures must be as low as −160°F (−110°C) and usually
require a combination of external refrigeration and expansion. These plants
require a demethanizer to increase recovery rates and to strip methane from
the NGL.
Gas volumes and gas composition set the optimal plant configuration on
an economic basis. This combination makes it difficult to set criteria for
establishing the best plant configuration. However, the higher the gas volume
ethane recoveries.
Inlet gas pressures make a major difference in plant configuration. High
pressures permit use of expansion, J-T or turboexpander, to provide all of the
cooling if low ethane recovery is desired. For low inlet pressures, either external
refrigeration or inlet compression followed by expansion is needed to cool the
gas, regardless of extent of ethane recovery. Required outlet pressure helps decide
which approach should be taken.
and GPM (gal liquid per Mscf [see Chapter 1]), the more attractive are high
© 2006 by Taylor and Francis Group, LLC
184 Fundamentals of Natural Gas Processing
The following three sections discuss the three hydrocarbon-recovery systems:
1. Dew point control and fuel conditioning
2. Low ethane recovery
3. High ethane recovery
The focus is on simple configurations, to give the reader the ability to evaluate
and understand the more complex configurations they may see.
7.3.1 DEW POINT CONTROL AND FUEL CONDITIONING
Dew point control and fuel conditioning exist to knock out heavy hydrocarbons
from the gas stream. These operations are primarily field operations.
7.3.1.1 Low Temperature Separators
Low temperature separators (LTS) (also called low temperature extraction units,
or LTX) are used both onshore and offshore. The process consists of cooling and
partial condensation of the gas stream, followed by a low temperature separator.
When inlet pressures are high enough to meet discharge-pressure requirements
to make pressure drop acceptable, cooling is obtained by expansion through a J-T
valve or turboexpander. Otherwise, external cooling is required. Water usually is
present, and to prevent hydrate formation the separator downstream of the
expander is warmed above the hydrate-formation temperature to prevent plugging.
An alternative to heating is injection of either ethylene glycol or methanol, which
is then recovered and dried for reuse.
prevention and uses J-T expansion for cooling. The feed initially goes through
a water knockout vessel to remove free water. The water-saturated hydrocarbon
gas and liquids then mix with ethylene glycol before being precooled. The
mixture then passes through a J-T expansion valve and flashes into the low
temperature separator to separate the gas, condensate, and glycol−water
phases. The condensate goes to the condensate stabilizer for removal of
remaining light ends.
Overhead gas from the low temperature separator passes through the pre-
cooler before being combined with the stabilizer overhead and put into the
pipeline. The low temperature separator is set to maintain the proper dew point
of the blended outlet gas.
The C
3
+ condensate from the stabilizer goes to product storage. The glycol−
water mixture from the low temperature separator goes to the glycol regenerator
for removal of the water and then reinjection into the feed.
If inlet pressures are too low for expansion, the stream is cooled by propane
refrigeration. The advantage of direct refrigeration is that the pressure drop is kept
at a minimum. Hydrate formation must be considered with either feed dehydration
Figure 7.8 shows a LTS that uses ethylene glycol injection for hydrate
© 2006 by Taylor and Francis Group, LLC
Hydrocarbon Recovery 185
upstream of the unit or inhibitor injection. Glycol injection is usually the more
cost effective, but if used, it increases the required refrigeration duty. The Engi-
neering Data Book (2004e) provides additional details on these processes.
7.3.1.2 Twister
Okimoto and Betting (1997) reported use of a relatively new device, Twister, for
dew point control and dehydration. It is used in one offshore facility (Brouwer
et al, 2004) and has been favorably evaluated for other projects (Wilson and
salient parts of the unit. Gas enters and expands through a nozzle at sonic velocity,
which drops both the temperature and pressure and causes droplet nucleation.
The two-phase mixture then contacts a wing that creates a swirl and forces
separation of the phases by centrifugal force. The gas and liquid are separated in
the diffuser; the liquid is collected at the walls and dry gas exits in the center.
Advantages of the system include:
• Simplicity. No moving parts and no utilities required.
• Small size and low weight. A 1-inch (24-mm) throat diameter, 6 feet
(2 m) long tube can process 35 MMscfd (1 MMSm
3
/d) at 1,450 psia
(100 bar).
• Driven by pressure ratio, not absolute pressure.
• Relatively low overall pressure drop. System recovers 65 to 80% of
original pressure.
• High isentropic efficiency. Efficiency is around 90% compared with
75 to 85% for turboexpanders.
FIGURE 7.8 Low-temperature separator (LTS), with glycol injection and condensate
stabilization.
Condensate
stabilizer
Glycol
regenerator
Condensate
Rich glycol
Low
temperature
separator
J-T
valve
Water
Gas from field
Water
knockout
Residue gas
Yuvancic, 2004). Figure 7.9 shows a cutaway view of the device and denotes the
© 2006 by Taylor and Francis Group, LLC
186 Fundamentals of Natural Gas Processing
Drawbacks of the system include:
• Requires a clean feed. Solids erode the tubing and wing, necessitating
an inlet filter separator.
• Limited turndown capacity. Flow variability is limited to ±10% of designed
flow. This limitation is mitigated by use of multiple tubes in parallel.
Liquids exit with a “slip gas.” This mixture is typically 20 to 30% of the total
flow volume (Epsom, 2005). The mixture can go to a gas−liquid separator for
recovery of the gas, which may require recompression.
7.3.1.3 Vortex Tube
Vortex tubes use pressure drop to cool the gas phase but generate both a cold
and warm gas stream. If streams are recombined, the overall effect is comparable
to a J-T expansion. The principle of operation is the Ranque-Hilsch tube, devel-
oped in the 1940s and commonly marketed as a means to provide cold air from
a compressed air stream. Cockerill (1998) provides an excellent general and
fundamental discussion of the device. Lorey and Thomas (2005) discuss the
principles of the device and provide operating data on its use for gas conditioning.
For dew point control, and dehydration, the device has the vortex tube and a
liquid receiver connected to the tube. Gas enters the tube tangentially through several
nozzles at one end of the tube, expands, and travels spirally at near sonic velocities
to the other end. As it travels down the tube, warm and cool gas separate. The cool
gas goes into the center of the tube. Warm gas vents in a radial direction at the end,
but the cool gas is reflected back up the tube and exits just beyond the inlet nozzles.
Condensation occurs in the cool gas, and the liquid is moved to the walls by centrifugal
force, where it collects and drains into the receiver below. The overall cooling effect
is comparable to that of a J-T expansion, with a low-temperature separator. However,
the vortex tube combines the expansion and separation into a single step.
The working pressure of the tube is 500 to 3,050 psig (36 to 210 barg), and
flow rates are 20 to 140 MNm
3
/h. The turndown ratio is 15% for a single tube
FIGURE 7.9 Cutaway view of Twister device. (Courtesy of Twister BV.)
Saturated
inlet gas
Laval nozzle Supersonic wing
Liquids and
slip gas
Dry gas
Diffuser
Expander Cyclone separator Compressor
© 2006 by Taylor and Francis Group, LLC
Hydrocarbon Recovery 187
but can be increased by use of multiple tubes in parallel; the optimum pressure
drop is 25 to 35% (Lorey and Thomas, 2005). The vendor states that liquid
condensation must be less than 10 wt%. (FILTAN, undated). Lorey and Thomas
(2005) note that the device performs well with up to 5% liquids in the inlet stream.
The device has been used to dehydrate gas from underground storage. To prevent
hydrate formation in the cold stream, TEG is added.
Like Twister, the vortex tube has the advantage of simplicity and light weight.
It could be useful where limited turndown is acceptable. It will be of most value
when no compression is required.
7.3.1.4 Membranes
cessing, including dew pointing. Hale and Lokhandwala (2004) discuss use of mem-
branes for fuel conditioning. Membranes are ideal for this application, provided
preconditioning is adequate to protect the membrane, and little penalty exists for
permeate compression. Figure 7.10 shows the flow configuration. Gas enters the
membrane on the discharge side of the compressor, and the residual gas provides fuel
to the compressor engine or turbine. The low pressure permeate is recycled to the
field unit. Gas rates are low because only a slip stream needs to be processed for fuel.
Like the previous two technologies, the process is simple and requires no moving
parts. It too has the advantage of being relatively small and light weight. (The
membrane unit described by Hale and Lokhandwala [2004] is 6 ft × 6 ft × 6 ft [2
m × 2 m × 2 m]). The authors report this technology is used on several offshore
installations. Unlike the Twister and vortex tube, membranes have the advantage of
a turndown ratio down to 50%, with no performance penalty. This property may not
be an advantage for fuel gas conditioning, where flow rates should be stable.
The reported data are used to compute the permeate composition and fraction
of feed removed from the feed. The table points out the selectivity of the membrane
FIGURE 7.10 Schematic for membrane unit used as a fuel conditioner. (Adapted from
Hale and Lokhandwala, 2004.)
Rich gas
Compressor Cooler
Lean fuel
gas
Membrane
Compressor
engine
Fuel gas
slipstream
Enriched gas permeate
suction for recompression to recover the permeate. Table 7.5 provides results for one
As discussed in Chapter 5, membranes are being used in several areas of gas pro-
© 2006 by Taylor and Francis Group, LLC
188 Fundamentals of Natural Gas Processing
and may be poorer than that of the above two technologies. In fuel conditioning,
the selectivity is not a major issue because of the relatively small fraction of gas
that needs to be recompressed, and the enriched stream is recycled without
requiring additional compression.
A membrane that passes heavy hydrocarbons preferentially to smaller mol-
the product of the solubility and diffusion coefficient. For separation of light
gases, the primary mode of selectivity is the diffusion coefficient. For dew point-
ing, solubility drives the selectivity (Baker et al., 1998). These membranes are
silicone rubber compounds that preferentially absorb the heavy components.
7.3.2 LOW ETHANE RECOVERY
The focus of the previous section was removal of heavy components (C
3
+) to
avoid condensation or to lower the heating value. This section discusses processes
TABLE 7.5
Operating Conditions and Composition of Natural Gas
Stream Using Membrane for Fuel Gas Conditioning
Membrane
Feed
Conditioned
Fuel Gas Permeate
Temperature, °F (ºC) 95 (35) 51 (10.5) —
Pressure psig (barg) 940 (65) 940 (65.0) —
Total mass flow lb mol/h
(kg-mol/h)
110.1 (50) 58.0 (26.3) 52.1 (23.7)
Total volume flow
MMscfd (MSm
3
/d)
0.95 (27) 0.5 (14) —
Mol%
% Removed
from Gas Component Feed Fuel Gas Permeate
a
Carbon dioxide 1.3 0.6 2.08 76
Methane 72.8 81.2 63.59 41
Ethane 9.6 9.0 10.29 51
Propane 9.9 7.1 13.04 62
i-Butane 2.4 0.8 4.19 82
n-Butane 2.5 0.9 4.29 81
n-Pentane 1.3 0.4 2.30 84
Water 0.11 0.00 0.23 100
Hydrocarbon dew
point (ºC)
35 3.5 —
a
Composition of permeate and fraction removed from gas computed by material balance of nor-
malized reported feed and fuel gas composition.
Source: Hale and Lokhandwala (2004).
ecules is counterintuitive. Chapter 5 points out that membrane permeability is
© 2006 by Taylor and Francis Group, LLC
Hydrocarbon Recovery 189
used in conventional gas plants, where the objective is to produce a lean gas and
recover up to approximately 60% of the ethane in the feed gas. Two process
schemes are used to obtain this level of ethane recovery:
• Cooling by expansion or external refrigeration
• Lean-oil absorption
As noted above, inlet pressure dictates the best means of refrigeration. Lean
oil was an early method used for hydrocarbon recovery but is now used on a
more limited basis. Many of the refrigerated lean oil absorption plants in operation
today are large facilities, where replacing them with a more modern turboex-
pander plant would be capital cost prohibitive. Both approaches are described
briefly below.
7.3.2.1 Cooling by Expansion or External Refrigeration
Figure 7.11 shows the propane recovery obtainable as a function of process temper-
ature for various levels of liquids content. In this case, GPM is based upon the C
3
+
GPM based upon C
3
+ fraction. Both the figures point out how recovery depends
upon the relative amount of C
3
+ in the gas stream and the lowest gas temperature.
A general rule is to assume that recovery increases with increased richness of the
gas. This assumption is made because the ethane content in the vapor at the top of
FIGURE 7.11 Recovery of the propane plus fraction as a function of the lowest separation
temperature and gas composition of feed. Operating pressure is 600 psig, and numbers given
on chart represent GPM based upon C
3
+. (Adapted from Engineering Data Book, 2004e.)
20
30
40
50
60
70
80
90
−40 −30 −20 −10 0 10 20
Process temperature, °F
C
3

r
e
c
o
v
e
r
y
,

%
3
5
7
fraction. Figure 7.12 shows a similar plot for recovery of ethane and propane, with
© 2006 by Taylor and Francis Group, LLC
190 Fundamentals of Natural Gas Processing
the column is set by column feed composition, along with temperature and pressure.
At constant pressure and temperature, the ethane concentration in the liquid decreases
with increasing C
3
+ fraction, which lowers the ethane concentration in the vapor
and, thus, increases the percent ethane recovered. (However, this outcome will not
always be the case in plants that use J-T or turboexpanders, because leaner gas puts
less of a load on the refrigeration-expander system and may lower column temper-
atures and increase recovery (McCartney, 2005)).
recycle from a fractionator to maximize liquids recovery. Inlet gas is initially cooled
with cold residue gas and cold liquid from the cold separator before going to the
propane chiller and to the cold separator. Vapor from the separator is the sales gas,
and the liquid goes to a fractionator to strip out light ends and recover liquid product.
The column operates at a lower pressure than does the cold separator. Because of
system pressure drop and because the fractionator runs at the lower pressure, the
recycle stream must be recompressed. Alternatives to the process include:
• Reduction or elimination of the recycle by adding reflux to the fractionator
• Running the fractionator at a higher pressure and use of a pump to
feed the column from the cold separator
FIGURE 7.12 Recovery of ethane and propane as a function of the lowest separation
temperature and gas content of feed. Operating pressure is 600 psig and GPM are based
upon C
3
+. The solid and dashed lines represent ethane and propane recovery, respectively.
(Adapted from Engineering Data Book, 2004e.)
10
20
30
40
50
60
70
80
90
−30 −25 −20 −15 −10 −5 0 5 10
Process temperature, °F
R
e
c
o
v
e
r
y
,

%
3 GPM
3 GPM
5 GPM
7 GPM
5 GPM
7 GPM
Figure 7.13 shows one commonly used direct-refrigeration process that employs
© 2006 by Taylor and Francis Group, LLC
Hydrocarbon Recovery 191
The Engineering Data Book (2004e) provides more details of process varia-
tions and discusses how processing variables affect compressor and refrigeration
requirements. These configurations assume that the gas enters sufficiently dehy-
drated to prevent hydrate formation. If the water content is higher, ethylene glycol
can be added, which increases refrigeration duty. However, temperatures then are
limited by glycol viscosity.
Because the unit relies only on external propane refrigeration, the lower
temperature limit on the feed to the cold separator is −35°F (−37°C) at best.
Unless the feed has a very high GPM, ethane recoveries will be below 60%.
Expansion is required to lower temperatures and increase recoveries.
With high inlet gas pressures, replacing the propane system with an expander
is an attractive option. However, inlet compression may be necessary to obtain
the temperatures required to obtain the desired recoveries. Both J-T and turboex-
panders are used. Crum (1981) points out situations where a J-T system may be
preferable to turboexpanders, although recent advances in turboexpander tech-
nology may temper some of them:
• Low gas rates. J-T is more economically viable at low gas rates. Crum
(1981) maintains that at below 10 MMscfd (300 MSm
3
/d), turboex-
panders offer less economic advantage and they lose efficiency below
5 MMScfd (150 MSm
3
/d).
• Low ethane recovery. For ethane recoveries of 10 to 30%, J-T expan-
sion may be sufficient.
• Variable flow rates. J-T is insensitive to widely varying flow rates, whereas
turboexpanders lose efficiency when operating off of design rates.
FIGURE 7.13 Schematic of a direct refrigeration process for partial recovery of C
2
+
fraction. (Adapted from Engineering Data Book, 2004e.)
Cold separator
Propane
chiller
Fractionator
Compressor Air cooler
Inlet gas
NGL product
Reboiler
Residue
gas
© 2006 by Taylor and Francis Group, LLC
192 Fundamentals of Natural Gas Processing
Crum (1981) also points out that J-T plants are much simpler than turboex-
pander plants because J-T plants have no need for seal gas and lubricating oil
systems. However, because of the inefficiency of J-T valves compared with
turboexpanders, if any inlet compression is required, more is required with J-T
expansion to obtain the same amount of refrigeration. The Engineering Data Book
(2004e) suggests that use of J-T expansion for limited ethane recovery requires
inlet pressures around 1,000 psi (70 bar).
Crum (1981) discusses five configurations in which J-T expansion is used to
recover C
2
+ in plants, ranging from 3 to 10 MMScfd (80 to 300 MSm
3
/d). Without
external refrigeration, recovery rates were up to 39%. A process that utilizes
propane refrigeration in combination with J-T expansion obtained 80% recovery.
Crum (1981) found that the fuel cost savings for compression failed to justify
installation of turboexpanders in these small plants.
7.3.2.2 Lean Oil Absorption
Early gas processing plants used lean oil absorbers to strip NGL from natural
gas (Cannon, 1993), and the process is still used in about 70 gas plants today.
To improve recoveries, later plants used external refrigeration to cool the feed
gas and lean oil. Figure 7.14 shows a representative schematic of a propane-
refrigerated lean oil system. The process involves three steps (Engineering Data
Book, 2004e):
1. Absorption. An absorber contacts a lean oil to absorb C
2
+ plus from
raw natural gas.
FIGURE 7.14 Refrigerated lean oil absorption process. (Adapted from Engineering Data
Book, 2004e.)
Inlet gas
Residue gas
Absorber Rich oil
demethanizer
Still
NGL
Fuel gas
Propane
chiller
Propane
chiller
© 2006 by Taylor and Francis Group, LLC
Hydrocarbon Recovery 193
2. Stabilization. The rich oil demethanizer (ROD) strips methane and
lighter components from the rich oil.
3. Separation. The still separates the recovered NGL components as prod-
uct from the rich oil, and the lean oil then returns to the absorber.
Gas from the ROD is either blended with the exiting gas stream or used for
fuel. Original systems used lean oil with molar mass of 150 to 200, but refrigerated
systems use molar masses of 100 to 130.
If no refrigeration is used, and assuming the absorber runs at about 100°F
(38°C), over 75% of the butanes and essentially all of the C
5
+ fraction are
recovered. Using high solvent rates makes possible the recovery of 50% of the
ethane and essentially all of the propane and heavier components (McCartney,
2005). With propane refrigeration, typically over 97% of the propane is recovered
and up to 50% of the ethane. The Engineering Data Book (2004e) notes that
refrigeration of the inlet gas and lean oil, along with heat to the still and ROD,
are the key elements of an efficient lean oil system.
An advantage of lean oil absorption is that little pressure drop occurs through
the absorber. However, the process is energy intensive and relies on numerous
heat exchangers to reduce the energy load. For gas processing, the whole process
can be simplified by elimination of the lean oil and use of external refrigeration,
as discussed in the previous section. However, many refrigerated lean oil absorp-
tion plants remain in operation today with capacities of 1,000 MMscfd (30
MMSm
3
/d) or more (McCartney, 2005).
One use for lean oil absorbers today is in capturing fugitive hydrocarbons
from air streams because refrigeration is unnecessary. Many non gas related
industries use this process for pollution control.
7.3.3 HIGH ETHANE RECOVERY
The above processes provided limited recovery of ethane. To obtain 80 to 90%
or more ethane recovery requires separation temperatures well below what is
obtainable by use of propane refrigeration alone. In principle, direct-refrigeration
processes could be used by cascading propane cooling with ethane or ethylene
refrigeration or by use of a mixed refrigerant that contains methane, ethane, and
propane. The primary motivation for use of only direct refrigeration would be
low inlet gas pressures. If significant inlet compression is required to produce
refrigeration by expansion, then cascade or mixed-refrigeration cooling, with or
without expansion, may be attractive. No matter which option is used, obtaining
high ethane recoveries from low inlet-pressure feed streams requires substantial
compression, of either the feed stream, the refrigerants, or both.
With recovery of a high ethane fraction, sales gas specifications must be consid-
ered. Recovery of too much ethane could reduce the heating value below contract
(37,000 kJ/m
3
) heating value as a function of GPM on the basis of C
2
+ and inerts
(e.g., N
2
, CO
2
) in the sales gas. The plot assumes that all inerts in the inlet gas remain
limits. Figure 7.15 shows the maximum recovery attainable to obtain a 1,000 Btu/scf
© 2006 by Taylor and Francis Group, LLC
194 Fundamentals of Natural Gas Processing
in the sales gas. Figure 7.16 shows how ethane recovery affects propane and butane
recoveries. The Engineering Data Book (2004e) notes that the propane recovery
varies, depending upon process configuration.
consists of a gas−gas heat exchanger with five gas streams that enter at different
FIGURE 7.15 Maximum ethane recovery as a function of noncombustible concentration
in the feed and ethane plus concentration. Numbers denote the GPM (based upon C
2
+) in
the feed. Lines assume that all inerts remain in the sales gas, and the sales gas heating
value is 1,000 Btu/scf. (Adapted from Engineering Data Book, 2004e.)
FIGURE 7.16 Recovery of propane and butanes as a function of ethane recovery rate.
The solid and dashed lines represent propane and butanes recovery, respectively. (Adapted
from Engineering Data Book, 2004e.)
30
40
50
60
70
80
90
100
0 2 4 6 8 10 12 14
Nonhydrocarbon content, mole %
M
a
x
i
m
u
m

e
t
h
a
n
e

r
e
c
o
v
e
r
y
,

%
1.5
3.0
5.0 8.0
85
90
95
100
30 40 50 60 70 80 90 100
Ethane recovery, %
R
e
c
o
v
e
r
y
,

%

Figure 7.17 shows a simplified conventional expander plant schematic. It
© 2006 by Taylor and Francis Group, LLC
Hydrocarbon Recovery 195
temperatures, cold separator, turboexpander, and demethanizer. Although the flow
sheet shown is schematically simple, in practice most actual designs replace the
single exchanger with a more complex and efficient combination of exchangers.
The inlet gas stream makes several passes through the gas−gas exchanger before
going to the cold separator, where the vapor expands through a turboexpander.
Liquid from the cold separator is flashed through a J-T valve and fed to the middle
of the demethanizer. The incoming gas provides reboiler heat at the bottom, and
then is cooled further in a second reboiler midway up the column.
A J-T valve is always installed parallel to the turboexpander. This configu-
ration helps in plant start-up and in handling excess gas flow. It also is used if
the turboexpander goes down.
The maximum ethane recovery with the conventional turboexpander config-
uration is about 80% (Engineering Data Book, 2004e). Also, the cold separator
may be near the critical temperature and pressure of the mixture, which can make
the process unstable. Carbon dioxide freezing out can also be a problem. Improve-
ment of C
2
+ recovery requires reduction of ethane losses in the top of the
demethanizer by addition of reflux. The Engineering Data Book (2004e) discusses
a number of configurations used. One that can provide up to 98% recovery, called
maximum ethane recovery with regards to compression requirements of all com-
monly used processes. It has the added advantage that it can reject ethane and
still maximize C
3
+ recovery if desired.
FIGURE 7.17 Schematic of conventional turboexpander process with no recycle to
demethanizer. Note that the one heat exchanger represents a network of exchangers.
(Adapted from Engineering Data Book, 2004e.)
NGL
Residue gas to
outlet compression
Cold propane
liquid
Feed gas
Cold propane
vapor
Demethanizer
Cold
separator
Turboexpander
Reboiler
Side reboiler
J-T valve
By pass J-T
valve
the cold residue recycle (CRR) process, is shown in Figure 7.18, which gives the
© 2006 by Taylor and Francis Group, LLC
196 Fundamentals of Natural Gas Processing
In this variation, the cold separator runs at a warmer temperature to avoid
the critical point problem. The vapor from the separator splits into two streams.
Part goes to the turboexpander and the balance goes through two overhead
exchangers, where it is condensed to provide liquid reflux to the column. Tur-
boexpander output then enters further down the column. In addition, part of the
overhead is compressed and cooled to provide additional reflux.
The Engineering Data Book (2004e) provides details on a number of other
process variations that are commonly used.
7.4 SAFETY AND ENVIRONMENTAL
CONSIDERATIONS
Safety considerations primarily exist because of the high pressures and low
temperatures involved in hydrocarbon recovery. Plugging from insufficient
removal of water, BTEX, and carbon dioxide can be a processing hazard. Properly
sized relief systems must be properly functioning in case of plugging.
Environmental concerns are relatively minor; lubricating oils along with
methanol or glycol, if used, are the primary potential pollutants from spills. All
process fluids are tied directly to the flare in case of emergency shutdown.
FIGURE 7.18 Cold-residue recycle process for maximizing ethane recovery. All valves
in figure are J-T expander valves but are unlabeled for figure clarity and the large heat
exchanger represents a network of exchangers. (Adapted from Engineering Data Book,
2004e.) 16-21
NGL
Residue gas to
outlet compression
Propane refrigerant
Feed gas
Cold propane
vapor
Demethanizer
Cold
separator
Turbo-
expander
Reboiler
Side
reboiler
© 2006 by Taylor and Francis Group, LLC
Hydrocarbon Recovery 197
REFERENCES
Agahi, R. and Ershaghi, B., Turboexpander Redesign Concepts and Economics, Proceed-
ings of the Seventy-First Annual Convention of the Gas Processors Association,
Tulsa, OK, 1992, 29.
Baker, R.W, Wijmans, J.G, and Kaschemekat, J.H., The design of membrane vapor-gas
separation systems, J. Membrane Sci., 151, 55, 1998.
Brouwer, J.M., Bakker, G., Verschoof, H.-J., and Epsom, H.D., Supersonic Gas Condi-
tioning First Commercial Offshore Experience, Proceedings of the Eighty-Third
Annual Convention of the Gas Processors Association, Tulsa, OK, 2004.
Cannon, R. E., The Gas Processing Industry, Origins and Evolution, Gas Processors
Association, Tulsa, OK, 1993, 51.
Crum, F.S., Application of J-T Plants for LP-Gas Recovery, Proceedings of the Sixtieth
Annual Convention of the Gas Processors Association, Tulsa, OK, 1981, 68.
Cockerill, T.T., Thermodynamics and Fluid Mechanics of a Ranque-Hilsch Vortex Tube.
Engineering Data Book, 12th ed., Sec. 2, Product Specifications, Gas Processors Supply
Association, Tulsa, OK, 2004a.
Engineering Data Book, 12th ed., Sec. 9, Heat Exchangers, Gas Processors Supply Asso-
ciation, Tulsa, OK, 2004b.
Engineering Data Book, 12th ed., Sec. 13, Compressors and Expanders, Gas Processors
Supply Association, Tulsa, OK, 2004c.
Engineering Data Book, 12th ed., Sec. 14, Refrigeration, Gas Processors Supply Associ-
ation, Tulsa, OK, 2004d.
Engineering Data Book, 12th ed., Sec. 16, Hydrocarbon Recovery, Gas Processors Supply
Association, Tulsa, OK, 2004e.
Engineering Data Book, 12th ed., Sec. 19, Fractionation and Absorption, Gas Processors
Supply Association, Tulsa, OK, 2004f.
Epsom, H., private communication, 2005.
Hale, P. and Lokhandwala, K., Advances in Membrane Materials Provide New Solutions
in the Gas Business, Proceedings of the Eighty-Third Annual Convention of the
Gas Processors Association, Tulsa, OK, 2004.
Haynes, B.S. and Johnson, A.M., High-Effectiveness Micro-Exchanger Performance, pre-
Herrin, J.P., New process for liquid recovery, Hydrocarbon Process., 45 (6) 144 1966.
Howard, I., Hannibal’s Experiences, Proceedings of the Laurance Reid Gas Conditioning
Conference, Norman, OK, 1998, 194.
Jumonville, J., Tutorial on Cryogenic Turboexpanders, Proceedings of the Thirty-Third
Lillard, J.K., and Nicoll, G., An Operating History of Turboexpanders in the Supercritical
Fluid Regime, Proceedings of the Seventy-Third Annual Convention of the Gas
Processors Association, Tulsa, OK, 1994, 270.
Lorey, M. and Thomas, K., Gas Conditioning Utilizing the Vortisep Device Application
of a Novel Technology on an Industrial Scale, Proceedings of the Eighty-Fourth
Annual Convention of the Gas Processors Association, Tulsa, OK, 2005.
Turbomachinery Symposium, 2004, http://turbolab.tamu.edu/pubs/Turbo33/T33
pg127.pdf, Retrieved September 2005.
Ph.D. thesis, University of Cambridge, Cambridge, England, 1998, www.south-
street.freeserve.co.uk/rhvtmatl/, Retrieved July 2005.
sented at AIChE 2002, Spring National Meeting, New Orleans, www.heat-
ric.com/technical_papers.html, Retrieved September 2005.
FILTRAN GmbH, http://www.filtan.de/ENGLISH/VTS_A.htm, Retrieved June 2005.
© 2006 by Taylor and Francis Group, LLC
198 Fundamentals of Natural Gas Processing
Lunsford, K.M., Understand the use of brazed heat exchangers, Chem. Eng. Progress, 92
(11) 44, 1996.
McCartney, G.D., private communication, 2005.
Okimoto, F.T, and Betting, M., Twister Supersonic Separator, Proceedings of the Laurance
Reid Gas Conditioning Conference, Norman OK, 2001.
Peranteaux, J., private communication, 2005.
Pua, L.M. and Rumbold, S.O., Industrial Microchannel Devices—Where Are We Today?
First International Conference on Microchannels and Minichannels, Rochester,
New York, 2003.
Sell, K.T., Langston, P.R. and Mitchell, R.H., Compressor Station Fuel Gas Superheating
Using Lube Oil Waste Heat, International Pipeline Conference, October 4−8,
Calgary, Alberta, Canada, 2004.
Shah, K. and Stucky, B., Operation of High Pressure Cryogenic Demethanizerater El Paso
Eunice Gas Plant, Proceedings of the Eighty-Third Annual Convention of the Gas
Processors Association, Tulsa, OK, 2004.
Smith, J.M., Van Ness, H.C., and Abbott, M.M., Introduction to Chemical Engineering
Thermodynamics, 6th ed., McGraw-Hill, New York, 2001.
Wadekar, V.V., Compact heat exchangers, Chem. Eng. Progress., 96 (12) 39, 2000.
Wilson, J. L. and Yuvancic, J., Process Selection for Dehydrating Gulf of Mexico Offshore
Platform Gas, Proceedings of the Laurance Reid Gas Conditioning Conference,
Norman, OK, 2004, 125.
© 2006 by Taylor and Francis Group, LLC
199
8
Nitrogen Rejection
8.1 INTRODUCTION
Three sets of circumstances require nitrogen separation or rejection:
• Processing a gas high in nitrogen to produce a pipeline quality gas.
• Removing nitrogen from a natural gas so that the nitrogen can be used
in an enhanced oil recovery (EOR) operation.
• Separating helium from nitrogen in a helium recovery operation.
Field Liquids Removal
Gas & Liquids from Wells
Inlet
Compression
Sulfur
Recovery
Gas Treating
Dehydration
Nitrogen
Rejection
Helium
Recovery
Outlet
Compression
Elemental
Sulfur
Field Dehydration
Field Acid Gas Removal
Field Compression
Sales Gas
Water &
Solids
Natural
Gas Liquids
Natural
Gasoline
Raw Helium
CO
2
Liquefaction
Liquefied
Natural Gas
Inlet Receiving
Hydrocarbon
Recovery
Liquids Processing
© 2006 by Taylor and Francis Group, LLC
200 Fundamentals of Natural Gas Processing
This chapter discusses nitrogen rejection for EOR and for production of pipeline
2000) found that approximately 24 Tcf (670 Bm
3
), 16% of the nonassociated reserves
at that time, were subquality in nitrogen and, consequently, will require blending to
lower the nitrogen concentration or processing to remove the nitrogen to meet the 3
mol% total inerts specification for pipelines (Engineering Data Book, 2004a).
In 1998 (EPRI, 1999), enhanced oil recovery (EOR) methods contributed about
12% of the total oil production in the United States. Of the EOR production, about
55% was from thermal methods, 28% was from carbon dioxide flooding, 12% was
from natural gas flooding, and 4.5% was from nitrogen flooding. Even with the
relatively small fraction of production, nitrogen EOR contributed 32,000 barrels of
oil per day (5,100 m
3
/d).
The inlet nitrogen concentration to a plant upgrading subquality gas is rela-
tively constant, although it may fluctuate if variations exist in nitrogen concen-
trations in the various gases feeding the plant. This circumstance is not the case
in EOR applications. In nitrogen EOR projects, nitrogen is injected into a stim-
ulation well to increase oil production in producing wells. Gas from the producing
wells initially contains little or no nitrogen. As the nitrogen gradually breaks
through from the injection to the producing wells, the nitrogen concentration
gradually rises until it reaches a value so high that the project is terminated.
Consequently, in EOR applications the nitrogen rejection unit (NRU) must be
designed to accommodate changing inlet feed concentrations. This chapter briefly
discusses both upgrading subquality gas and EOR applications.
Tannehill et al. (1999) discuss the processing of high nitrogen gas. They also
cover the locations of high nitrogen reserves and the number and types of plants
used in the processing, and they summarize some of the economics.
8.2 NITROGEN REJECTION FOR GAS UPGRADING
Three basic methods are used for removal of nitrogen from natural gas:
• Cryogenic distillation
• Adsorption
• Membrane separation
are the most economical and can provide higher nitrogen rejection at high gas
throughput. At low gas volumes, membranes and pressure swing adsorption (PSA)
by use of molecular sieves are economically feasible. The tabulated flow ranges
are guidelines only.
In regard to hydrocarbon recovery, only PSA has heavier hydrocarbons (all
C
4
+ and part of propane) going with the nitrogen stream. This situation is caused
by adsorption in the sieve binder, as the components are too large to enter the
sieve pores. The binder also adsorbs water and CO
2
. The loss of hydrocarbons
may or may not be beneficial.
Table 8.1 provides a comparison of the three processes. Cryogenic methods
quality gas. A brief discussion of the helium−nitrogen separation is left to Chapter 9.
As mentioned in Chapter 1, the Gas Research Institute’s survey of reserves (Meyer,
© 2006 by Taylor and Francis Group, LLC
Nitrogen Rejection 201
8.2.1 CRYOGENIC DISTILLATION
The most common method of removing nitrogen from natural gas is cryogenic
distillation. The Engineering Data Book (2004b) notes that for feed concentrations
below 20% N
2
, a single-column design can be used. For higher concentrations,
a dual-column is better. With the addition of a recycle compressor, it can be used
at lower N
2
contents. Figure 8.1 shows a flow diagram for a two-column NRU
TABLE 8.1
Comparison of Nitrogen Removal Processes
Process
Flow Range
MMscfd
(MSm
3
/d) Complexity
Heavy
Hydrocarbon
Recovery
Development
Stage
Cryogenic
distillation
>15
(400)
Complex In product gas Mature
Pressure-swing
adsorption
(PSA)
2 − 15
(60 − 400)
Simple; batch operation,
requires bed switching
In regeneration
gas
Early
commercialization
Membrane 0.5 − 25
(15 − 700)
Simple continuous
operation
In product gas Early
commercialization
Source: Adapted from Hale and Lokhandwala (2004).
FIGURE 8.1 NRU by use of two-column cryogenic distillation (Handwerk, 1990). Valves
are J-T valves.
N
2
−260°F 10 psig
98% C
1
−200°F
10 psig
65% N
2
35% C
1

2
0
0
°
F

1
6
0
°
F
Demethanizer
overhead
15% N
2
−150°F, 250 psig
−210°F
−175°F
200 psig
−245°F
15 psig

2
4
0
°
F

3
0
0
°
F

2
5
0
°
F
© 2006 by Taylor and Francis Group, LLC
202 Fundamentals of Natural Gas Processing
receiving feed that contains 15% N
2
from a demethanizer in a conventional
turboexpander plant. Gas from the demethanizer overhead is cooled by heat
exchange and pressure reduction and fed to a distillation column operating at
200 psig (14 barg). The bottoms product from this high-pressure column is
reduced in pressure to cool the stream to −240°F (−151°C). This stream, combined
with the bottoms product from the second low-pressure column, is fed to a heat
exchanger in the top of the high-pressure column to provide the necessary reflux.
The overhead from the high-pressure column flows through three heat exchangers,
is reduced in pressure to approximately 15 psig (1 barg), and enters the low-
pressure column at −300°F (−184°C). The overhead from this column is 98% N
2
,
and the bottoms product is approximately 98% CH
4
. The Hannibal Gas Plant of
British Gas Tunisia (Jones et al., 1999) uses cryogenic distillation to reduce the
N
2
content of the feed gas from 16.9% N
2
to the sales-gas specification of 6.5%.
operations are given by Jones et al. (1999) and Howard (1998).
8.2.2 PRESSURE SWING ADSORPTION
After cryogenic distillation, pressure swing adsorption (PSA) is probably the
most widely used process. At this point, we should briefly discuss the significant
differences between the adsorption process used for dehydration (thermal swing
adsorption, or TSA) and that used for nitrogen rejection (pressure swing adsorp-
tion, or PSA).
the means of separation in adsorption. The amount adsorbed depends on four
factors:
1. The adsorbent itself
2. The species being adsorbed (adsorbate)
3. The temperature
4. The partial pressure of the adsorbate
Once the adsorbent and adsorbates are selected, the temperature and partial
pressure become the governing variables. All industrial regenerative adsorption
separations involve two steps: adsorption to separate the species, followed by
desorption and removal of the adsorbate (regeneration) to prepare the adsorbent
for further use.
In natural gas systems, if adsorption is used to remove a relatively small
amount of material to a very low level, or if the heat of adsorption is very high,
TSA is generally used, as discussed in Chapter 6. An example is natural gas
dehydration, which meets both criteria. For bulk removal of one component from
another (e.g., upgrading natural gas to pipeline specifications by removal of CO
2
or N
2
), PSA may be the choice because concentrations of the adsorbate are high
and the heat of adsorption is low. We briefly discuss the fundamentals of this
process.
(See Chapter 15 for more details.) Complete descriptions of the plant and its
As discussed in Chapter 6, differences in adsorbate polarity and size provide
© 2006 by Taylor and Francis Group, LLC
Nitrogen Rejection 203
Figure 8.2 shows a very simplified two-bed PSA system (actual plants may
have four beds) to separate a binary mixture of compounds A and B. Assume A
is strongly adsorbed relative to B. The feed enters the adsorbing bed at ambient
temperature and a high pressure with the partial pressure of A = p
1
. Because A
is more strongly adsorbed than B, the adsorbed phase (adsorbate) is enriched in
A, and the A content of the gas leaving the bed is reduced. When the outlet
concentration of A begins to increase because of bed breakthrough, it is switched
to the regeneration mode, and the feed is switched to the previously regenerated
bed to continue the cycle.
Regeneration is accomplished by dropping bed pressure, which causes the gas
to desorb, and purging the bed to remove the desorbed gas. The adsorption isotherm
in Figure 8.2 (the plot of weight fraction A, X
A
, adsorbed versus partial pressure)
indicates that the final concentration of A on the regenerated bed drops to what will
be in equilibrium with the purge gas A at a partial pressure of p
2
. Thus, all of the
A cannot be removed from the bed. The residual A that loads when the bed is put
back into adsorption service is at equilibrium with p
2
.

This condition means the gas
that leaves the bed during the adsorption mode has a partial pressure of p
2
. If the
purge gas contains no A, then its

partial pressure is reduced to zero, and most of
the adsorbed A will be desorbed and purged. Consequently, the bed is capable of
reducing the level of A in the product to a very low level.
This brief discussion is very simplified. Readers who desire a more complete
treatment should consult the review article of Sherman and Yan (1991).
D’Amico et al. (1993) describe a PSA unit that uses a specially treated carbon
molecular sieve (CMS) for nitrogen rejection. The CMS is prepared by processing
of coal or wood carbon. Complete details as well as an economic study are presented
in the original article. Mitariten (2001) and Mitaritan and Dolan (2001) describe
a PSA process, called Molecular Gate
TM
, that uses a new type of molecular
FIGURE 8.2 Simple pressure swing adsorption (PSA) system.
Feed
Partial pressure
of A = p
1
Purge gas
Partial pressure of A = p
2
Product gas
Partial pressure
of A = p
2
W
e
i
g
h
t

f
r
a
c
t
i
o
n

o
f

c
o
m
p
o
n
e
n
t

A

a
d
s
o
r
b
e
d
Partial pressure
p
2
p
1
X
A1
Adsorption mode
High pressure
Ambient temperature
Regeneration mode
Low pressure
Ambient temperature
X
A2
© 2006 by Taylor and Francis Group, LLC
204 Fundamentals of Natural Gas Processing
sieve. According to these authors, conventional molecular sieve 3A and 4A have pore
openings of 3.2 and slightly less than 4.2 Å, respectively. The new Molecular Gate
sieve is a titanium silicate with a pore opening of 3.7 Å, which allows better separation
of nitrogen (3.6 in diameter) from methane (3.8 in diameter). Further information on
the Molecular Gate process is given by Mitaritan et al (2002).
8.2.3 MEMBRANES
sion applies to the present problem of N
2
separation from natural gas. Hale and
Lokhandwala (2004) discuss the NitroSep™ process for removal of N
2
from
natural gas to produce a pipeline quality gas along with a fuel gas (Figure 8.3).
In the process shown, the low-pressure inlet gas is compressed, any condensate
formed is removed, and the feed goes to the membrane unit, where pipeline-
quality gas and a liquid stream that contains most of the heavier hydrocarbons
are produced. The residue gas from the first membrane unit goes to a second unit
that produces a gas suitable for use as a compressor fuel and a recycle stream to
recover the methane.
8.3 NITROGEN REJECTION FOR ENHANCED
OIL RECOVERY
As discussed previously, the primary difference between conventional nitrogen
rejection and nitrogen rejection for EOR is the planned increase with time in N
2
feed concentration to the NRU for EOR. The reason for the wide variation lies
in the nature of the EOR process. The N
2
is injected into the reservoir through a
series of injection wells positioned to force the oil to the producing wells. Initially,
little or no N
2
will be in the gas and oil produced, but inevitably, N
2
begins to
break through and reach the producing wells. As the field becomes depleted, the
N
2
content increases. Eventually, the N
2
level reaches a point at which gas
production is no longer economically possible, and the project is stopped.
In the EOR project described by McLeod and Schaak (1986), the anticipated
inlet NRU N
2
concentrations ranged from 21 to 75%. As previously noted, feed
FIGURE 8.3 Separating N
2
from natural gas by use of membranes.
Recycle gas (10–15% N
2
)
Low-pressure gas
(10% N
2
)
Membrane
module
Membrane
module
Condensate
Pipeline gas (<4% N
2
)
Fuel gas
(50% N
2
)
Chapter 5 presented the basic features of membrane separation, and that discus-
© 2006 by Taylor and Francis Group, LLC
Nitrogen Rejection 205
concentration plays a major role in the design (Engineering Data Book, 2004b).
Therefore, these NRU plants must be carefully designed to accommodate a wide
range of inlet concentrations. Davis et al. (1983) discuss EOR designs in which
the N
2
level of the gas goes from 5 to 80% N
2
.
The process of choice for NRUs in EOR service is cryogenic distillation, and
the major decision is the type of distillation scheme, single column or double
column. Both processes are discussed by Davis et al. (1983); the single column is
presented in more detail by Davis et al. (1989). The double-column system shown
(1985). The double-column system can more easily handle large changes in N
2
and
is preferred for EOR or where the N
2
feed content can change widely.
McLeod and Schaack (1986) give a detailed account of an enhanced oil
recovery project that utilizes N
2
as the injection fluid. Figure 8.4 is a general
schematic of the operation, which consists of three separate units:
• An air separation plant to generate nitrogen
• A gas plant to recover liquids from the produced stream
• An NRU to produce a sales gas
The air separation plant produces 30 MMscfd (0.8 MMSm
3
/d) of N
2
that
contains 5 ppmv of O
2
, in addition to some liquid nitrogen. The oxygen from the
unit is vented to the atmosphere. The N
2
gas is compressed to 4,900 psig (340 barg)
in two sets of compressors and injected into the reservoir. The solution gas that comes
FIGURE 8.4 Schematic of an enhanced oil recovery (EOR) system.
NRU
Gas plant
Air separation
plant
Air
O
2
LN
2
to sales
N
2
LPG
Sales gas
N
2
N
2
Reservoir
in Figure 8.1 and described above is described in more detail by Alvarez and Vines
© 2006 by Taylor and Francis Group, LLC
206 Fundamentals of Natural Gas Processing
from the reservoir is treated in a conventional gas processing plant to produce a
heavier hydrocarbon product (LP gas) and a residue gas that goes to the NRU.
The NRU must produce a sales gas with a heating value of at least 975 Btu/scf
(36,300 kJ/m
3
) from a feed gas that contains from 20 to 75% N
2
. The NRU uses
double-column cryogenic distillation for the separation, which is described in the
paper. At the beginning of the project little N
2
is present in the produced gas, but
the concentration continuously increases during the life of the project. In the later
stages of the project, N
2
removed in the NRU is recompressed and reinjected into
the reservoir. This procedure reduces the need for N
2
from the air separation plant.
The Anschutz Ranch East NGL/NRU plant described by Davis et al. (1989)
was designed to process 300 MMscfd (8 MMSm
3
/d) of gas with N
2
compositions
varying from 2 to 80%. When built, it was the largest NRU plant in the world to
process widely varying N
2
feeds. Additional information on NRUs may be found
in the papers of Goethe and Mawer (1987) and Young and Maloney (1989).
8.4 SAFETY AND ENVIRONMENTAL
CONSIDERATIONS
Safety considerations are centered on the high pressures and extremely low
temperatures generally encountered in NRUs. In addition, if the low-temperature
nitrogen can be vented, it must be routed to a safe location so that it becomes
more buoyant than the ambient air and is not allowed to accumulate in a confined
space. This procedure prevents the possibility of an accidental asphyxiation.
Environmentally, all of the systems are benign.
REFERENCES
Alvarez, M.A. and Vines, H.L., Nitrogen rejection/NGL recovery for EOR projects, Energy
Prog., 5 (2) 67 1985.
Davis, R.A., Vines, H.L., and Pervier, J.W., Cryogenic Schemes for Nitrogen Rejection—
an EOR Project, presented at the Rocky Mountain Meeting of the Gas Processors
Association, 1983.
Davis, R.A., Gottier, G.M., Blackburn, G.A., and Root, C.R., Integrated Nitrogen Rejection
Facility Design and Operation, Proceedings of the Sixty-Eighth Annual Conven-
tion of the Gas Processors Association, Tulsa, OK, 1989, 204.
D’Amico, J., Reinhold, H., and Gamez, J., A PSA Process for Nitrogen Rejection from
Natural Gas, Proceedings of the Laurance Reid Gas Conditioning Conference,
Norman, OK, 1993, 55.
Engineering Data Book, 12th ed., Sec. 2, Product Specifications, Gas Processors Supply
Association, Tulsa, OK, 2004a,.
Engineering Data Book, 12th ed., Sec. 16, Hydrocarbon Recovery, Gas Processors Supply
Association, Tulsa, OK, 2004b.
EPRI, Enhanced Oil Recovery Scoping Study, TR-113836, Palo Alto, CA. 1999.
2005.
www.energy.ca.gov/process/pubs/electrotech_opps_tr113836.pdf, Retrieved June
© 2006 by Taylor and Francis Group, LLC
Nitrogen Rejection 207
Goethe, A. and. Mawer, D.J, A New Integrated Nitrogen Rejection Process with NGL
Recovery, Proceedings of the Sixty-Sixth Annual Convention of the Gas Processors
Association, Tulsa, OK, 1987, 89.
Hale, P. and Lokhandwala, K., Advances in Membrane Materials Provide New Gas Pro-
cessing Solutions, Proceedings of the Laurance Reid Gas Conditioning Conference,
Norman, OK, 2004, 165.
Handwerk, G., private communication, 1990.
Howard, I., Hannibal’s Experiences, Proceedings of the Laurance Reid Gas Conditioning
Conference Norman, OK, 1998, 194.
Jones, S., Lee, S., Evans, E., and Chen, R., Simultaneous Removal of Water and BTEX
From Feed Gas for a Cryogenic Plant, Proceedings of the Seventy-Eighth Annual
Convention of the Gas Processors Association, Tulsa, OK, 1999, 108.
McLeod, R.P. and Schaak, J.- P., A Unique Combination of Natural Gas Processing with
Cryogenic Technology in an Enhanced Oil Recovery Project, Proceeding of the
Sixty-Fifth Annual Convention of the Gas Processors Association, Tulsa, OK,
1986, 71.
Meyer, H.S., Volume and distribution of subquality natural gas in the United States, Gas
TIPS 6, 10, 2000.
Mitariten, M., New technology improves nitrogen removal economics, Oil Gas J., 99 (17)
42, 2001.
Mitaritan, M., and Dolan, W., Nitrogen Removal from Natural Gas with Molecular Gate
Technology, Proceedings of the Laurance Reid Gas Conditioning Conference,
Norman, OK, 2001.
Mitaritan, M., Dolan, W., and Maglio, A., Innovative Molecular Gate Systems for Nitrogen
Rejection, Carbon Dioxide Removal, and NGL Recovery, Proceedings of the
Eighty-First Annual Convention of the Gas Processors Association, Tulsa, OK,
2002.
Sherman, J.D. and Yan, C.M., Adsorption, gas separation, in Encyclopedia of Chemical
Technology, 4th ed., Wiley-Interscience, New York, 1991.
Tannehill, C.C., Strickland, J.G., and Meyer, H., High N
2
Gas—Snap Shot of the Present
Requirements for the Future, Proceedings of the Seventy-Eighth Annual Convention
of the Gas Processors Association, Tulsa, OK, 1999, 230.
Young, S.M. and Maloney, J.J., Enhanced Reservoir Value Through Nitrogen Rejection,
Proceedings of the Laurance Reid Gas Conditioning Conference, Norman, OK,
1989, 301.
© 2006 by Taylor and Francis Group, LLC
209
9
Trace Component
Recovery or Removal
9.1 INTRODUCTION
impurities. A number of other trace components present processing, product
quality, or environmental problems if concentrations are too high. These compo-
nents include:
Field Liquids Removal
Gas & Liquids from Wells
Inlet
Compression
Sulfur
Recovery
Gas Treating
Dehydration
Nitrogen
Rejection
Helium
Recovery
Outlet
Compression
Elemental
Sulfur
Field Dehydration
Field Acid Gas Removal
Field Compression
Sales Gas
Water &
Solids
Natural
Gas Liquids
Natural
Gasoline
Raw Helium
CO
2
Liquefaction
Liquefied
Natural Gas
Inlet Receiving
Hydrocarbon
Recovery
Liquids Processing
Chapters 5, 6, 7, and 8 discussed methods to remove the common natural gas
© 2006 by Taylor and Francis Group, LLC
210 Fundamentals of Natural Gas Processing
• Hydrogen
• Oxygen
• Radon (NORM)
• Arsenic
• Helium
• Mercury
• BTEX (benzene, toluene, ethylbenzene, and xylene)
This chapter briefly touches on the first four items in the list and then discusses
the last three items in more detail, primarily because they are of such importance
that several commercial processes exist to remove them.
9.1.1 HYDROGEN
Hydrogen levels are rarely sufficiently high to cause a problem, and no GPA
specifications exist for maximum levels in sales gas. The primary source
would be gas streams from refineries, but even there, the concentrations should
be low.
9.1.2 OXYGEN
Of the items listed above, oxygen is the only contaminant that is not naturally
occurring, and the best approach for treating it is to prevent its introduction into
the processing stream. The major source of oxygen is leaking valves and piping
in gathering systems that operate below atmospheric pressure. The Engineering
Data Book (2004b) gives a 1.0 vol% maximum oxygen concentration in sales
gas, but some pipelines have specifications as low as 0.1 vol%. However, it causes
problems in gas processing at concentrations of 50 ppmv, such as:
• Enhancing pipeline corrosion if liquid water is present
• Reacting with amines in gas treating, which ultimately leads to heat stable
salt formation
• Reacting with glycols to form corrosive acidic compounds
• Reacting with hydrocarbons during the high-temperature regeneration
of the adsorption beds to form water, which reduces regeneration
effectiveness and, thus, reduces bed capacity
At low levels, the oxygen can be removed by use of nonregenerative
scavengers. For higher concentrations, the obvious removal method is to cata-
lytically react it in the gas stream to produce water, which is then removed in
the dehydration step. However, gases that contain sulfur compounds poison oxi-
dation catalysts. (The motivation for reduction of sulfur levels in gasoline and
diesel fuels.) Thus, combustion upstream of gas treating is infeasible unless the
gas is H
2
S free. Oxidizing the process stream after treating generates CO
2
, which
can freeze out in cryogenic sections of the plant. Oxygen is removed in sales gas
with nitrogen rejection, but this step follows the processing steps most susceptible
© 2006 by Taylor and Francis Group, LLC
Trace Component Recovery or Removal 211
to oxygen damage. Also, few gas plants have nitrogen rejection units. Oxygen
will become more of a problem as gas fields in the United States are depleted.
9.1.3 RADON (NORM)
Natural gas contains radon, a naturally occurring radioactive material (NORM),
at low concentrations, and it rarely poses a health problem because it has a half-
life of about 3.8 days (Encyclopedia Americana, 1979). However, radon decays
into lead-210, then to bismuth-210 and polonium-210 and finally into stable
lead-206. These daughter products of radon, some of which have long half-lives,
condense on pipe walls and form a low-level radioactive scale, which may flake
off and collect on inlet filters. Because the boiling point of radon is −79.2°F
(−61.8°C), it tends to concentrate in propane and ethane−propane mixtures.
Storage vessels can accumulate the daughter products as sludge. Discarded piping
with the scale generates large quantities of low-level radioactive waste that must
be discarded in disposal wells. For further details see, Bland (2002), Gray (1990),
Grice (1991), and Railroad Commission of Texas (undated).
9.1.4 ARSENIC
Arsenic is a toxic nonvolatile solid but exists in natural gas predominately as a
more volatile trimethylarsine (As (CH
3
)
3
). It usually collects as a fine gray dust.
High concentrations tend to be geographically localized. It can be successfully
removed from gas by use of a nonregenerative adsorption process. Several facil-
ities reduce arsenic concentrations in sweet raw gas from around 1,000 to less
than 1µg/m
3
(Rhodes, 2005). Without arsenic removal, the gas streams could not
be marketed. The process requires dehydration of the gas to pipeline specifications
before it goes to the adsorbers.
9.2 HELIUM
9.2.1 INTRODUCTION
Helium is a difficult diluent to remove in natural gas unless nitrogen rejection
is used. Unlike the other trace components discussed here, helium is a valuable
product from natural gas processing, which makes high concentrations desir-
able. Recovery of helium from air is technically feasible but would be very
expensive because the helium concentration is so low, 5.24 ± 0.05 ppm (Engi-
neering Data Book, 2004a). The only viable source is from select natural gas
fields, and even the best of these gas fields contain relatively small quantities
of helium. A helium-rich gas is defined as one that contains more than 0.3%
helium, and according to published field concentrations (Deaton and Haynes,
1961; Guccione, 1963; Johnson and Rydjord, 2001), a gas that contains more
than 5% is not usually found.
The National Academy of Sciences (2000) presents an excellent overview of
helium reserves and usage in the United States. In 2003, the world production of
© 2006 by Taylor and Francis Group, LLC
212 Fundamentals of Natural Gas Processing
grade-A helium (99.995% purity) was 6,400 MMscf (181 MMm
3
); the United
States produced 84% of the total, and the remainder coming from Algeria, Poland,
and Russia. However, two new large helium plants are in the planning and
construction stages, one in Darwin, Australia, slated to start up in 2007, and the
other in Qatar, which becomes operational in 2005. Both plants will process the
offgas from baseload LNG facilities. At full production, the Darwin plant will
be capable of producing 150 MMscf (4.2 MMm
3
) of helium per year (BOC
Group, 2005) and the Qatar plant will be capable of producing 300 MMscf
(8.5 MMm
3
) per year (Pacheco, 2003).
The uses of helium are limited but important. Its main applications stem from
the fact that it is chemically inert and has the lowest boiling point of any substance.
Figure 9.1 shows how helium was used in the United States in 2003.
9.2.2 RECOVERY METHODS
Recovery of helium from natural gas requires refinement of the low concentrations
into high-purity helium (99.995%). However, the gas plant may produce only a
raw helium stream that contains a roughly equimolar mixture of helium and
nitrogen. This stream then goes to another facility for final purification. Early
processing schemes (Deaton and Haynes, 1961; Guccione, 1963; Remirez, 1968)
were cryogenic and involved cooling and condensing the natural gas and then
recovering the helium by distillation as a crude product that was about 65 to 80%
helium. The crude helium was then stored or further refined in a series of low-
temperature separations, followed by a final adsorption step on activated charcoal
to produce the high-purity product.
FIGURE 9.1 Helium uses in the United States in 2003 (Pacheco, 2003).
Cryogenics, 28%
Leak
detection,
4%
Pressure or
purge, 26%
Welding, 20%
Other, 7%
Controlled
atmosphere, 13%
Breathing
mixtures, 2%
© 2006 by Taylor and Francis Group, LLC
Trace Component Recovery or Removal 213
More recently, interest has increased in alternate technologies such as pressure-
swing adsorption (PSA) and membranes for helium recovery. Choe et al. (1988)
present a discussion and analysis of hybrid cryogenic and membrane processes.
The newest helium recovery plant in the United States, the Ladder Creek
Plant (Johnson and Rydjord, 2001), located in Eastern Colorado, started produc-
tion in 1998. Figure 9.2 is a simplified process flow diagram for the Ladder Creek
are representative of the feed gases to the plant.
The plant was designed to process 35 MMscfd (1 MMm
3
/d), with expansion
capability to 50 MMscfd (1.4 MMm
3
/d). Helium recovery was set at 95%, with
the ethane either rejected or recovered. The natural gas liquids (NGL) are recovered
while pipeline gas is produced.
In Figure 9.2, the carbon dioxide content is first reduced to less than 10 ppm,
and then trace quantities of mercury present are removed before dehydration by
use of molecular sieve beds. Details of the cold box are unavailable, but its
products are:
• Crude helium stream (nitrogen plus helium)
• Small stream of liquid nitrogen
• Residue gas
• Natural gas liquids
The liquid nitrogen is used to fill the nitrogen shields on the liquid helium
trailers and the helium storage. The crude helium is purified to 99.997% by use of
FIGURE 9.2 Schematic of Ladder Creek Helium Recovery Plant (Johnson and Rydjord,
2001).
Feed gas from
gathering system
Waste gas
recycle
compressor
Liquid nitrogen
storage
Helium
storage
Truck filling
station
Tube trailer
filling station
High pressure
storage
NGL/NRU/HRU
cold box
PSA
helium
purifier
Helium
liquefier
Molecular sieve
dehydration
Carbon dioxide
removal
Crude helium
pretreatment
Helium product
compressor
Nitrogen to vent
or liquid storage
Residue gas
NGL product
Plant. The inlet gas composition varies substantially, but the values in Table 9.1
pressure swing adsorption (PSA). (See Chapter 8 for details on PSA.) The PSA
© 2006 by Taylor and Francis Group, LLC
214 Fundamentals of Natural Gas Processing
adsorbers operate at ambient temperature and 240 to 280 psig (16.5 to 19.3 barg).
Adsorbers are on-line for approximately 320 seconds and are then taken off-line,
depressurized to approximately 40 psig (3.8 barg), and flushed with pure helium
gas for reactivation. All flush gas is recycled to the cold box. The helium produced
may be liquefied and stored and shipped by truck as a liquid or compressed and
stored and shipped as a high-pressure gas. Details of the patented helium liquefier
are not available, but it has a capacity of 1.7 MMscfd (50 MSm
3
/d).
Example 9.1 Compute the ratio of the minimum work of separation required to
obtain pure He from air compared with that from a natural gas that contains 3 mol%
He in the feed. Assume the separation takes place at constant temperature and that
the gases are ideal.
As shown in any chemical engineering thermodynamics textbook (e.g., Smith et al,
2001), the theoretical minimum amount of work to go from one state to another is
T∆S, where T is the absolute temperature and ∆S is the change in entropy between
the initial and final state. The entropy change per mole of ideal gas mixture formed
from pure components is
∆S = −RΣ x
i
ln x
i
,
where R is the gas constant and x
i
is the mole fraction of component i. Because the
He concentration is so low in air, the mole fraction of the other components in
air can be assumed to not change, so the change in entropy is assumed to be
caused by the change in He concentration only. We are interested in the ratio of
TABLE 9.1
Range of Feed Gas Compositions to Ladder Creek
Helium Recovery Plant
Composition (mol%)
Component Rich Helium Gas Lean Helium Gas
Nitrogen 61.92 31.58
Helium 3.54 1.81
Carbon dioxide 0.98 0.91
Methane 26.65 52.84
Ethane 2.60 6.40
Propane 1.95 3.64
i-Butane 0.37 0.46
n-Butane 0.88 1.09
Pentane+ 1.10 1.27
Total 100.00 100.00
Source: Johnson and Rydjord (2001).
© 2006 by Taylor and Francis Group, LLC
Trace Component Recovery or Removal 215
work per mole of He produced and base work on 1 mole of initial gas mixture.
Thus, the ratio of work is
If the two separations are performed at 100% thermodynamic efficiency at a constant
temperature, the work required to obtain helium from air would be 2.9 times greater
than extracting it from natural gas.
9.3 MERCURY
Two major problems are associated with the presence of mercury in natural gas:
amalgam formation with aluminum and environmental pollution. According to
Rios et al. (1998) mercury may be present in elemental form or as organometallic
compounds, such as dimethylmercury, methylethylmercury, and diethylmercury,
or as inorganic compounds such as HgCl
2
.
Elemental mercury distributes between the gas and liquid phases. Whereas
elemental mercury has an extremely low vapor pressure (~4 × 10
6
psia at 32°F,
3 × 10
−7
bar at 0°C, [Rios, et al. 1998]), vapor phase concentrations are orders
of magnitude higher than computed by Equation 6.1 because of the nonideal
behavior of mercury at operating conditions. For example, Stiltner (2002) found
gas from a high-pressure inlet separator contained about 25% of the mercury that
entered the plant. The high-pressure separator liquid was flashed to ambient
pressure, and the liquid contained less than 20% of the inlet mercury, whereas
the flash gas, which was mostly ethane through pentane, contained over 50% of
the inlet mercury. Thus, the majority of mercury goes to treating, dehydration,
and then to the cryogenic section. What fraction would reach the cryogenic section
was not stated.
The volatility of elemental mercury and the fact that it will accumulate as it
condenses makes removal from the gas stream mandatory in cryogenic plants.
Mercury corrodes brazed-aluminum heat exchangers as it amalgamates with the
aluminum to weaken the material. The first reported instance of this problem
occurred in an Algerian LNG facility in 1974 (Rios et al., 1998). Elemental
mercury is toxic to humans as well as a poison to many catalysts.
The mercury compounds concentrate in the hydrocarbon liquids, where they
potentially can present environmental and safety hazards. The compounds are
readily absorbed by most biological systems.
T S
T S
x
x x x
Air
NG
He Air
He NG He NG He


=
+ −
ln
ln (
,
, , ,
1
NNG He NG
He NG
x
x
) ln( )
ln( )
.
,
,
1
5 10
0 03
6







=
×

lln . . ln( . )
.
.
0 03 0 97 0 97
0 03
2 9
+






=
© 2006 by Taylor and Francis Group, LLC
216 Fundamentals of Natural Gas Processing
Although mercury in natural gas is normally at low levels, some gases contain
sufficiently high mercury concentrations to cause both safety and health concerns.
Table 9.2 shows the range of mercury levels in wellhead gas in a number of countries.
Abbott and Openshaw (2002) state that much higher mercury levels have
been detected in individual wells and that the highest known concentration is
4,400 µg/Nm
3
(500 ppbv) from a well in Germany. Rios et al. (1998) note that
Southeast Asian gases tend to have the higher elemental mercury levels, whereas
the United States Gulf Coast gases are usually low, but a wide variation can occur
within a given region.
9.3.1 ENVIRONMENTAL CONSIDERATIONS
The environmental considerations of mercury and its compounds fall into two
categories: (1) direct harm to individuals from inhalation of mercury vapors or
compounds and (2) safe disposal of contaminated materials and equipment.
From a health perspective, ingested mercury, particularly dimethyl mercury,
permanently damages the brain and kidneys. Unfortunately it concentrates in the
tissue of both the fish and shellfish that are consumed by humans. According to
the EPA database on fish advisories (Environmental Protection Agency, 2005a),
in 2005 the United States had 2,470 active advisories on eating fish and shellfish
because they contained high mercury levels. Stringent emission controls at the
federal level on mercury from coal-fired power plants are being considered that
will cut mercury emissions from power plants by 70% by 2018 (Environmental
Protection Agency, 2005b.).
The mercury concentration in most natural gases is insignificant compared with
the coals presently used in power plants (Environmental Protection Agency, 2001).
As discussed below, elemental mercury is removed from natural gas to protect pro-
cessing equipment and catalysts. However, some of these processes do not remove the
TABLE 9.2
Range of Elemental Mercury Levels in Wellhead
Natural Gas from a Number of Countries
Location
Elemental Mercury Concentration
in µg/Nm
3
(ppbv)
South America 69 – 119 (8 to 13)
Far East 58 – 93 (6 to 10)
North Africa 0.3 – 130 (0.03 to 15)
Gronigen (Germany) 180 (20)
Middle East 1 – 9 (0.1 to 1)
Eastern US Pipeline 0.019 − 0.44 (0.002 to 0.05)
Midwest US Pipeline 0.001 − 0.10 (0.0001 to 0.01)
North America 0.005 − 0.040 (0.0005 to 0.004)
Source: Abbott and Openshaw (2002).
© 2006 by Taylor and Francis Group, LLC
Trace Component Recovery or Removal 217
mercury compounds. The gas industry has not come under any EPA regulations that
require mercury removal. However, this situation could change in the future, espe-
cially for very large plants that process gases with higher than normal mercury levels.
The disposal of materials and equipment contaminated with mercury is a
complex issue, and a discussion is beyond the scope of this book. Rios et al.
(1998) and Müssig (1997) present excellent discussions of the problem.
9.3.2 AMALGAM FORMATION
The elemental form of mercury is apparently the cause of low temperature failure
in the commonly used plate-fin aluminum heat exchangers (Rios et al., 1998).
However, elemental mercury attacks copper, zinc and brass, chromium, iron, and
nickel as well. If the natural gas is to undergo cryogenic processing, the mercury
must be removed to levels below 0.01 µg/Nm
3
to avoid damage to aluminum heat
exchangers used in low temperature service. A detailed discussion of the mechanism
of mercury attack on brazed-aluminum exchangers is given by Nelson (1994).
9.3.3 REMOVAL PROCESSES
Both regenerative and nonregenerative processes are available for the removal of
mercury from gas and liquid streams, and both types will be discussed briefly. All
of the processes take advantage of the reactivity of elemental mercury for removal.
Complete removal of all mercury requires reduction of mercury compounds to
elemental mercury. Processes that use chemisorption for removal may simulta-
neously remove the organic mercury compounds through physical adsorption. Some
of the processes remove arsenic and lead as well (Engineering Data Book, 2004d).
9.3.3.1 Nonregenerative Processes
Rios et al. (1998) and Engineering Data Book (2004d) list a number of chemi-
sorption processes for removal of elemental mercury from hydrocarbon streams
to 0.01µg/Nm
3
or lower. Bed capacities of 10% or higher are common. Most of
these processes use sulfur impregnated on a support such as activated charcoal
or alumina to provide a large surface area. Bourke and Mazzoni (1989) describe
a sulfur-impregnated carbon adsorbent for mercury removal. In this process, the
mercury reacts with the sulfur to form a stable compound on the adsorbent surface.
The use of this type sulfur removal at the Anschutz Ranch East Plant in Wyoming
is discussed in some detail by Lund (1996).
Some processes do not adsorb the organic forms of mercury that are inert to
metals, and some treat only the gas phase.
9.3.3.2 Regenerative Process
Markovs and Corvini (1996) and Stiltner (2002) describe a regenerative pro-
cess that utilizes silver on molecular sieve to chemisorb elemental mercury
while providing dehydration at the same time. The advantage of silver is that
the amalgam it forms with mercury decomposes at typical regeneration temperatures
© 2006 by Taylor and Francis Group, LLC
218 Fundamentals of Natural Gas Processing
molecular sieve dehydration bed, and the basic dehydration process remains
unchanged. Essentially, all of the mercury condenses with the water on regeneration
and forms a separate phase, which easily can be decanted and sold. Stiltner
(2002) reports the solubility of water to be about 25 ppbw. Removal of the
mercury from the water requires an additional adsorption step. After water and
mercury removal, the regeneration gas can be recycled to the front of the bed
without additional treatment, or it can be blended into the sales gas. If neces-
sary, the regeneration gas can be further treated with a nonregenerative mer-
cury-removal process to remove the last traces of mercury.
Although the silver-sieve process treats both gas and hydrocarbon liquid, how
effective the process is in removing organic mercury compounds is unclear. Like
the nonregenerative processes, this process can drop elemental mercury levels to
0.01µg/Nm
3
.
9.4 (BTEX) BENZENE, TOLUENE, ETHYLBENZENE,
AND XYLENE
Benzene, toluene, ethylbenzene, and xylene (BTEX) present two possible prob-
lems in gas processing: environmental concerns and potential freezing in cryo-
genic units. In most gas processing plants, BTEX is present in the volatile organic
compounds (VOC) that must be controlled to meet EPA clean air regulations.
Benzene is also classified as a carcinogen under United States regulations and is
considered an air toxin. Emission limits for air toxics are lower than those for
VOCs. BTEX concentrates in the glycols and amines used to dehydrate and
sweeten the gas. Initially, industry interest was in glycol dehydration because it
posed the larger problem, but recently the amine sweetening process has been
considered as well.
In glycol dehydration, the glycol absorbs some hydrocarbons as well as
water. Although the paraffinic compounds are removed in the flash tank, a
fraction of the aromatic compounds are not completely removed because of their
relatively low vapor pressures at the flash conditions and the slight affinity of
glycol (the most commonly used glycol) to dehydrate a gas stream at 1,000 psia
(14.5 bar), 100°F (38°C) and at a typical glycol rate of 3 gal TEG/lb of water
absorbed, the TEG will absorb the following percentages of BTEX from the gas
stream (Engineering Data Book, 2004c):
When the rich glycol is regenerated, the BTEX exits the still in the vapor
stream, and consequently this stream must be processed to recover or eliminate
Benzene 10%
Toluene 14%
Ethyl benzene 19%
Ortho xylene 28%
for dehydration (see Chapter 6). The silver-impregnated sieve is added to the standard
glycols for them (see Chapter 6 for process description). Use of triethylene
© 2006 by Taylor and Francis Group, LLC
Trace Component Recovery or Removal 219
the BTEX before atmospheric venting. The methods presently used to control
BTEX emissions to the atmosphere are:
• Adjustment of plant operating conditions to minimize the quantity of
BTEX in the glycol absorber off gas
• Burning of the still off gases before venting
• Condensation of the off gases and recovery of the BTEX as a liquid
product
• Adsorption of the BTEX on a carbon adsorbent
BTEX removal is thoroughly discussed in the paper by Rueter et al. (1992),
and a summary of the main items in that paper is presented in the handbook by
Kohl and Nielsen (1997).
Amine treaters absorb BTEX along with the acid gases. Both acid gases and
the BTEX are stripped in the regenerator and, traditionally, either vented if the
H
2
S level is low or sent to a Claus unit for sulfur recovery. Collie et al. (1998)
present a comparative study of emissions from glycol dehydrators and amine
treaters. They used process simulation programs to model the behavior of both
glycol and amine units and, on the basis of simulation results, came to the
following conclusions:
• Reduction of the solvent circulation rate in both glycol and amine units
and minimization of the lean-amine temperature were the most effec-
tive means of reducing BTEX emissions from the flash gas.
• A change of solvent type can also reduce emissions.
• For the cases studied, approximately 25% of the BTEX emissions came
from amine units, and the remaining 75% came from glycol dehydrators.
Covington et al. (1998) present a similar paper that discusses how to minimize
both operating costs and BTEX emissions from glycol dehydrators by use of
ethylene glycol (EG), diethylene glycol (DEG), and triethylene glycol. They con-
clude that emissions can be reduced by use of either EG or DEG instead of TEG
because of the lower solubility of BTEX in the lighter glycols. Operating costs may
be reduced as well. However, this option may be attractive only when very low
water contents are not required and the cost of glycol losses is not excessive.
Like carbon dioxide, both benzene and p-xylene can cause plugging problems
in cryogenic NGL extraction units. If the feed contains significant quantities of
BTEX, these compounds must be removed to a level that will prevent freezing in
the low-temperature units. (GPA offers a computer program to estimate required
concentrations.) Jones et al. (1999) describe a plant that contains a nitrogen-rejec-
0.12% BTEX. The plant uses a glycol dehydrator (TEG) to remove water to a level
of 7 lb/MMscf and at the same time remove more than half of the BTEX. The
glycol unit is followed by a molecular sieve dehydrator to achieve the level of
dehydration necessary for the nitrogen rejection unit.
tion unit (see Chapter 15 for details) where the feed gas contains approximately
© 2006 by Taylor and Francis Group, LLC
220 Fundamentals of Natural Gas Processing
REFERENCES
Abbott, J. and Oppenshaw, P., Mercury Removal Technology and Its Applications, Pro-
ceedings of the Eighty-First Annual Convention of the Gas Processors Association,
Tulsa, OK, 2002.
Bourke, M.J. and Mazzoni, A.F., The Roles of Activated Carbon in Gas Conditioning,
Proceedings of the Laurance Reid Gas Conditioning Conference, Norman, OK,
1989, 137.
Choe, J.S., et al., Membrane/Cryogenic Hybrid Systems for Helium Purification, Proceed-
ings of the Sixty-Seventh Annual Convention of the Gas Processors Association,
Tulsa, OK, 1988, 251.
Collie, J., Hlavinka, M., and Ashworth, A., An Analysis of BTEX Emissions from Amine
Sweetening and Glycol Dehydration Facilities, Proceedings of the Laurance Reid
Gas Conditioning Conference, Norman, OK, 1998, 175.
Covington, K., Lyddon, L., and Ebeling, H., Reduce Emissions and Operating Costs with
Appropriate Glycol Selection, Proceedings of the Seventy-Seventh Annual Con-
vention of the Gas Processors Association, Tulsa, OK, 1998, 42.
Deaton, W.M. and Haynes, R.D., How newest helium plant is working, Pet. Refiner, 40,
(3) 205, 1961.
Encyclopedia Americana, Danbury, CT, 1979, 23, 130c.
Engineering Data Book, 12th ed., Sec. 1, General Information, Gas Processors Supply
Association, Tulsa, OK, 2004a.
Engineering Data Book, 12th ed., Sec. 2, Product Specifications, Gas Processors Supply
Association, Tulsa, OK, 2004b.
Engineering Data Book, 12th ed., Sec. 20, Dehydration, Gas Processors Supply Associ-
ation, Tulsa, OK, 2004c.
Engineering Data Book, 12th ed., Sec. 21, Hydrocarbon Treating, Gas Processors Supply
Association, Tulsa, OK, 2004d.
Environmental Protection Agency, Mercury in Petroleum and Natural Gas: Estimation of
Emissions from Production, Processing, and Combustion, EPA/600/R-01/066,
Retrieved July 2005.
Environmental Protection Agency, The National Listing of Fish Advisories (NLFA), 2005a,
Environmental Protection Agency, Fact Sheet—EPA Issues Notice of Data Availability on
Gray, P.R., Radioactive materials could pose problems for the gas industry, Oil Gas J., 88
(26) 45, 1990.
Grice, K.J., Naturally Occurring Radioactive Materials (NORM) in the Oil and Gas
Industry: A New Management Challenge, Presented at First International Confer-
ence on Health, Safety and Environment, The Hague, The Netherlands, Society
of Petroleum Engineers Paper SPE 23384, November 1991.
Guccione, E., New Approach to recovery of helium from natural gas, Chemical Eng., 70
(9) 73, 1963.
BOC Group, BOC to Build Australian Helium plant, 2005, http://www.boc.com/news/
article_866_09mar05.asp, Retrieved September 2005.
http://epa.gov/waterscience/fish/advisories/index.html, Retrieved October 2005.
Proposed Clean Air Mercury Rule, 2005b, http://www.epa.gov/mercury/control_
emissions/nodafact.html, Retrieved September 2005.
com/norm.htm, Retrieved July 2005.
Bland, C.J., A Review of NORM in Oil and Natural Gas Extraction, 2002, www.c5plus.
September 2001, www.epa.gov/ORD/NRMRL/pubs/600r01066/600r01066.htm,
© 2006 by Taylor and Francis Group, LLC
Trace Component Recovery or Removal 221
Johnson, D.J. and Rydjord, K., Unique NGL/NRU/HRU Plant Processes New Gas in
Eastern Colorado, Proceedings of the Eightieth Annual Convention of the Gas
Processors Assocation, Tulsa, OK, 2001.
Jones, S., Lee, S., Evans, M., and Chen, R., Simultaneous Removal of Water and BTEX
from Feed Gas for a Cryogenic Plant, Proceedings of the Seventy-Eighth Annual
Convention of the Gas Processors Association, Tulsa, OK, 1999, 108.
Kohl, A. and Nielsen, R., Gas Purification, 5th ed., Gulf Publishing, Houston, TX, 1997.
Lund, D.L., Causes and Remedies for Mercury Exposure to Aluminum Coldboxes, Pro-
ceedings of the Seventy-Fifth Annual Convention of the Gas Processors Associ-
ation, Tulsa, OK, 1996, 282.
Markovs, J. and Corvini, J., Mercury Removal from Natural Gas and Liquid Streams, Pro-
ceedings of the Laurance Reid Gas Conditioning Conference, Norman, OK, 1996, 133.
Müssig, S., Experience in Removing Mercury from Natural Gas and Subsequent Mercury
Decontamination of Process Equipment, Proceedings of the Laurance Reid Gas
Conditioning Conference, Norman, OK, 1997, 31.
National Academy of Sciences, The Impact of Selling the Federal Helium Reserve, National
Academies Press, Washington, D.C, 2000, books.nap.edu/catalog/9860.html,
Retrieved July 2005.
Nelson, D.R., Mercury Attack of Brazed Aluminum Heat Exchangers in Cryogenic Gas
Service, Proceedings of the Seventy-Third Annual Convention of the Gas Proces-
sors Association, Tulsa, OK 1994, 178.
Pacheco, N., Helium, U.S. Bureau of Land Management, Amarillo Field Office, Helium
Operations, Amarillo, TX, 2003.
Railroad Commission of Texas, NORM—Naturally Occurring Radioactive Material,
Remirez, R., Cold-box design is key in helium recovery unit, Chemical Eng., 75 (27)
80, 1968.
Rueter, C.O., et al., Research on Emissions of BTEX and VOC from Glycol Dehydrators,
Proceedings of the Laurance Reid Gas Conditioning Conference, Norman, OK,
1992, 187.
Rios, J.A., Coyle, D.A. Durr, C.A., and Frankie, B.M., Removal of Trace Mercury Con-
taminants from Gas and Liquid Streams in the LNG and Gas Processing Industry,
Proceedings of the Seventy-Seventh Annual Convention of the Gas Processors
Association, Tulsa, OK, 1998, 191.
Stiltner, J., Mercury Removal from Natural Gas and Liquid Streams, Proceedings of the Eighty-
First Annual Convention of the Gas Processors Association, Tulsa, OK, 2002.
Smith, J.M., Van Ness, H.C., and Abbott, M.M., Introduction to Chemical Engineering
Thermodynamics, 6th ed., McGraw-Hill, New York, 2001.
WEB SITES
This site provides information on a variety of environmental problems, including
mercury and hazardous wastes by organization of environmental officials in north-
eastern United States.
review of mercury and its health and environmental effects.
www.rrc.state.tx.us/divisions/og/key-programs/norm.html, Retrieved July 2005.
Rhodes, Z., private communication, 2005, www.newpointgas.com.
Northeast Waste Management Officials Association: www.newmoa.org/Newmoa/htdocs/
Pollution Probe: www.pollutionprobe.org/Publications/Mercury.htm. This site provides a
© 2006 by Taylor and Francis Group, LLC
223
10
Liquids Processing
10.1 INTRODUCTION
The previous chapters focused on processing gas streams. This chapter discusses
processing the hydrocarbon liquids. The two liquid streams processed in a gas
plant are the condensate from inlet separators and demethanizer bottoms NGL
stream. Condensate from field operations is brought by truck to the plant as well,
Field Liquids Removal
Gas & Liquids from Wells
Inlet
Compression
Sulfur
Recovery
Gas Treating
Dehydration
Nitrogen
Rejection
Helium
Recovery
Outlet
Compression
Elemental
Sulfur
Field Dehydration
Field Acid Gas Removal
Field Compression
Sales Gas
Water &
Solids
Natural
Gas Liquids
Natural
Gasoline
Raw Helium
CO
2
Liquefaction
Liquefied
Natural Gas
Inlet Receiving
Hydrocarbon
Recovery
Liquids Processing
© 2006 by Taylor and Francis Group, LLC
224 Fundamentals of Natural Gas Processing
but this condensate typically goes directly to storage and then to sale as natural
gasoline or slop oil without further processing. This chapter discusses the pro-
cessing required to produce marketable liquid products.
10.2 CONDENSATE PROCESSING
Figure 10.1 shows a schematic of one configuration for the condensate that leaves
the inlet receivers of a plant. Processing the condensate involves two steps: water
washing and condensate stabilization. Depending upon the associated water quality,
the condensate may require a water wash to remove salts and additives. If they are
not a problem, then incoming water is removed in the inlet receiver and a water
wash is unnecessary. The water often contains high concentrations of methanol or
ethylene glycol, added for hydrate prevention. Plants may recover the inhibitors
onsite or ship the water offsite for inhibitor recovery before disposal. In most cases,
waste water is injected into disposal wells (Kuchinski, 2005).
After removal of free water, the condensate goes to the stabilizer, where
remaining lighter hydrocarbons are stripped and recombined with the gas that
leaves the inlet receiver. The primary purpose of the stabilizer is to produce a
bottoms product that has specifications to be sold as “natural gasoline,” or “slop
oil.” Specifications on natural gasoline include:
• Volatility as measured by ASTM D323 (Standard Test Method for Vapor
Pressure of Petroleum Products (Reid Method)) test. The values range
from 10 to 34 psi (0.7 to 2.3 bar), but the common range is 9 to 12 psi
(0.6 to 0.8 bar), when the liquid is trucked offsite (Kuchinski, 2005). The
Engineering Data Book (2004a) lists the grades of natural gasoline on the
basis of RVP and percentage of liquid evaporated at 140°F (60°C).
FIGURE 10.1 Inlet receiver condensate-processing schematic.
Inlet Receiver
Gas and Liquids
from Field
Gas to Inlet Compression
Water to
Disposal
Storage
Natural
Gasoline
Water Wash
Stabilizer
Recycle
Compressor
Water
Makeup
© 2006 by Taylor and Francis Group, LLC
Liquids Processing 225
• Density set in the range of 65 to 85° API (650 to 720 kg/m
3
), with a
price penalty for higher gravities (lower densities).
• Water content, defined as BS&W (bottoms sediment and water), that
does not exceed 1 vol%, as determined by centrifuge. Higher values
discount sales price of natural gasoline. Langston (2003) provides
details on sampling and analysis for BS&W.

for details).
The Engineering Data Book (2004a) lists additional specifications related to
corrosion, distillation curve, color, and reactive sulfur.
The stripping column must operate at a temperature and pressure to ensure
that the bottoms product meets the volatility and density specification. It usually
operates in the 75 to 400 psig (5 to 27 barg) pressure range. Although lower
pressure reduces the heat duty on the stabilizer reboiler, the operating pressure
is usually dictated by the number of recycle compressor stages and available
capacity (Kuchinski, 2005). It usually is more economical to provide additional
reboiler duty than to add additional compression.
10.2.1 SWEETENING
Ideally, the sulfur compounds go with the light gases stripped in the stabilizer, so
that they can be removed in gas treating. Most of the H
2
S and CO
2
will be removed
along with COS, CS
2
, and mercaptans. (Mercaptans are commonly used odorants
in natural gas and propane.) To enhance acid gas removal, sweet natural gas can
be used as a stripping gas. This process can reduce H
2
S levels to the 10 ppmv range
(Webber et al., 1984). However, the cost is increased recycle compression.
If the natural gasoline is sour, the plant can either treat it or take a price
penalty and sell the liquid as sour crude. The price differential between sweet
and sour crude drives the need for sweetening. In many cases, the liquid volumes
are low and sweetening at the plant is not justified. Any of the sweetening
processes for NGL, discussed in detail below in 10.3.1, can be used. However,
some form of caustic wash is the most commonly used.
10.2.2 DEHYDRATION
Dehydration of natural gasoline usually is unnecessary. When in storage, free
water drops out and the remaining low water content should meet customer
requirements, provided the water layer is not shipped, too. However, emulsions
of water and natural gasoline occur, which require emulsion breakers to obtain
a clean water−hydrocarbon separation.
10.3 NGL PROCESSING
toms NGL stream, with extensive fractionation of NGL. The processing required
depends upon the plant’s feed quality and product slate and customer requirements.
Figure 10.2 shows a possible processing configuration for the demethanizer bot-
Reactive sulfur, as determined by the copper strip test (see Chapter 5
© 2006 by Taylor and Francis Group, LLC
226 Fundamentals of Natural Gas Processing
Processing demethanizer bottoms liquids ranges from no additional processing
to sweetening, dehydration and, extensive fractionation. Figure 10.2 implies that
all of the NGL is treated, which is not always the case. Feed quality and product
slate dictate which streams require treatment.
10.3.1 SWEETENING
As noted above, most of the sulfur compounds are stripped into the gas phase,
which, after any compression, typically goes to an amine treater. Ideally, all of
the sulfur compounds would be removed in the gas treating step. This step
eliminates additional sweetening and additional dehydration, if aqueous systems
extremely effective at removing the acid gases, but most do a poor job of removing
other sulfur compounds, especially mercaptans, sulfides, and disulfides. Also, the
acid gases and sulfur compounds concentrate in the demethanizer bottoms. There-
fore, extremely low sulfur levels are required in gas treating to meet the product
specifications in the liquid. If inlet gas contains both CO
2
and H
2
S, COS formed
COS present concentrates in the propane product stream (Mick, 1976). This
section discusses commonly used processes to remove sulfur compounds and
residual CO
2
from NGL.
FIGURE 10.2 NGL fractionation train.
NGL treating
Dehydration
Depropanizer
Debutanizer
Natural gasoline
to storage
Propane to
pipeline/storage
n-Butane to
pipeline/storage
i-Butane to
pipeline/storage
Demethanizer bottoms
Deethanizer
Ethane to
pipeline
(i.e., amines or caustics) are used. As noted in Chapter 5, gas treating units are
in regeneration of molecular sieve (see Chapter 6) may be present as well. Any
© 2006 by Taylor and Francis Group, LLC
Liquids Processing 227
10.3.1.1 Amine Treating
Amine treating of liquids is the same process used for treating gases. Most remove
2 2
but none remove CS
2
and mercaptans. Amine treating is often used upstream of caustic
treaters to minimize caustic consumption caused by irreversible reactions with CO
2
.
Amine treating is usually conducted in a packed tower, although sieve trays
are effective as well. To avoid vaporization, contactor operating pressure should
be 100 psi (7 bar) above the liquid bubble point, and the entire unit should operate
at least 50 psi (3.4 bar) above the bubble point (Nielsen, 1995). The Engineering
Data Book (2004d) recommends a minimum contactor packing depth of 20 feet
(6 m) and flow rates not exceeding 20 gpm per foot squared of cross-sectional
area (0.014 m
3
/s per m
2
).
Sargent and Seagraves (2003) and Veroba and Stewart (2003) provide addi-
tional details on contactor design. Nielsen (1995) notes that the lean-amine
viscosity should not exceed 2 cP to ensure good amine/hydrocarbon separation.
Even so, the expected normal concentration of amine in the hydrocarbon stream
that leaves the unit is about 100 ppmw. Water wash is recommended to both
recover amine and protect downstream caustic treaters. Veroba and Stewart (2003)
and Nielsen (1995) provide specifics on the water wash.
10.3.1.2 Adsorption
Adsorption is a commonly used method for sweetening NGL and LPG streams.
It has the advantage of being able to remove sulfur compounds down to low
levels when no water is present, which makes it attractive for treating demetha-
nizer bottoms. In contrast to the amine and caustic wash, no potential downstream
dehydration is required. Either promoted alumina, 13X or 5A molecular sieve
will remove H
2
S, COS, and mercaptans from LPG. The larger pore 13X is usually
preferred because it has a higher capacity (Richman, 2005). Mick (1976) discusses
the limitations of using molecular sieve for COS removal to the 2 ppm level when
concentrations are in the hundreds of ppm.
Hydrocarbon flow can be up or down during adsorption. However, upflow is
preferred, if possible, so that the regeneration can be down, to make recovery of
all liquid from the bed easier (Richman, 2005). Liquid velocities are usually 3
to 5 ft/min (0.9 to 1.5 m/s), and bed height should be at least 5 feet (1.5 m), to
allow good distribution (Engineering Data Book, 2004c). Liquid adsorption beds
use the same graduated support packing used in gas adsorption.
The bed must be drained of liquid, and a gas used for regeneration. If the
fluid being treated is butane or lighter, sweet vapor of the fluid being treated can
be used, which eliminates purging after regeneration. Refilling should be done
with dry sweet liquid. The filling rate must be slow to prevent bed movement.
Also, the bed needs to be cool to avoid flashing and bed movement (Richman,
2005). Bed rearrangement is detrimental because it induces channeling and causes
attrition; the generated dust leads to plugging in the bed or downstream filters.
only CO and H S. Two amines, DGA and DIPA (see Chapter 5), will remove COS,
© 2006 by Taylor and Francis Group, LLC
228 Fundamentals of Natural Gas Processing
The Engineering Data Book (2004c) indicates that new adsorbents capacity and
performance data are usually presented on the basis of static measurements. How-
ever, under flow conditions, the capacity may be about 50% of the static values,
depending on the total amount of adsorbent in the bed. This discrepancy is because
10.3.1.3 Caustic Treating
A number of caustic processes, both regenerative and nonregenerative, are used
to remove sulfur compounds from NGL. The simplest process is the use of a
nonregenerative solid KOH bed. It has the advantage of taking out H
2
S and
water as well. According to available literature, the nonregenerative solid KOH
process is not commonly used. Sambrook et al. (1997) provide some discussion
on the solid-bed KOH system. Mick (1976) found KOH effective for removal
of H
2
S but not other sulfur compounds. However, by injection of methanol into
the bed with the propane, COS, mercaptans, and CS
2
can be removed. The
process has the advantage of reducing sulfur content, while still producing a
dry product.
Eckersley and Kane (2004) suggest that nonregenerative metal oxides, such
as ZnO, are economically attractive if the sulfur rate into the bed is less than 400
lb/d (180 kg/d). The bed operates at 300 to 700°F (150 to 370°C).
One of the most common processes for treating natural gasoline, and NGL,
is the regenerative caustic wash with sodium hydroxide. Frequently, it follows
treating with an amine system. The amine pretreatment eliminates the acid gases
and reduces the load on the caustic wash that removes mercaptans.
The caustic wash process is similar to the hot potassium carbonate process
For H
2
S removal, the process is regenerable, but CO
2
forms a nonregenerative
carbonate salt. This by-product, along with the consumption of caustic for H
2
S
removal, is why the amine process precedes the caustic wash.
wash. The water wash eliminates carry over of the treating solution in the hydro-
carbon. Sand towers can be used to disengage the two liquid phases as well.
Hydrocarbon enters at the bottom of the contactor through a distributor to
ensure that it is the dispersed phase. The tower contains either packing or sieve
trays. The top 8 to 10 feet (2.4 to 3 m) of the contactor contains no packing or
trays to permit the phases to separate (Engineering Data Book 2004d).
Multiple stages of caustic wash are used if essentially complete removal is
required. For a single-stage unit, the caustic concentration is 10 to 15 wt%; if
multiple stages are used, the first stage concentration is 4 to 6 wt% and the others
are 8 to 10 wt% (Engineering Data Book 2004d). The contactor operates near
ambient temperature and at a sufficiently high pressure to avoid vaporization.
The regenerator bottoms temperature is usually around 275°F (135ºC) (Kuchinski,
2005). This system can be highly corrosive if not operated properly or if oxygen
is present in the stream being treated (McCartney, 2005).
Figure 10.3 shows the process for a single-stage treatment along with a water
of the mass transfer zone and bed aging (see Chapter 6 for details).
discussed in Chapter 5. It is extensively used in refineries for treating sour streams.
© 2006 by Taylor and Francis Group, LLC
Liquids Processing 229
The chemistry involved in the wash is straightforward:
RSH + NaOΗ ↔ RSNa + H
2
O (10.1)
H
2
S + NaOH ↔ NaSH + H
2
O (10.2)
NaSH + NaOH ↔ Na
2
S + H
2
O (10.3)
H
2
S + 2NaOH ↔ Na
2
S + 2H
2
O (10.4)
CS
2
+ 2NaOH → 2NaSH + CO
2
(10.5)
CO
2
+ 2NaOH → Na
2
CO
3
+ H
2
O (10.6)
The RSH in Equation 10.1 denotes a mercaptan, and their removal is a prime
reason for use of a caustic wash. Unfortunately, as Equation 10.5 shows, carbon
disulfide, which is usually at low concentrations compared with the other sulfur
compounds, reacts with sodium hydroxide to form CO
2
, which produces the
nonregenerative carbonate salt in Equation 10.6.
Some variations on the caustic wash to extract mercaptans exist. One process,
Merox (Verachtert et al., 1990), forms the water-soluble mercaptan salt in
FIGURE 10.3 Regenerative caustic wash and water wash.
NGL feed
Treated
NGL
Water wash
Mercaptans
Regenerator
Contactor
Coalescer
Water
makeup
Steam
Water to
disposal
Caustic wash
Filter
Filter Coalescer
Mixer
© 2006 by Taylor and Francis Group, LLC
230 Fundamentals of Natural Gas Processing
Equation 10.1. Air is added to the rich caustic, with a catalyst to oxidize the
mercaptan to a disulfide:
2RSNa + 1/2 O
2
+ H
2
O → RSSR + 2NaOH (10.7)
The disulfide is water insoluble and separated from the caustic before it
returns to the wash. The paper describes process alternatives that reduce the
required amount of caustic and treat different hydrocarbon streams. The separa-
tion of caustic and disulfide is somewhat difficult; although the phases are immis-
cible, the two liquids have almost identical densities (McCartney, 2005). Even
when treating streams with no CO
2
present, the Merox
TM
caustic will eventually
become spent from the CO
2
in the regeneration air stream.
The Engineering Data Book (2004d) describes a continuous process that
oxidizes the mercaptan by reducing cupric chloride to cuprous chloride. Both the
Merox
TM
and copper processes require the feed to be free of H
2
S.
10.3.1.4 Other Processes
The focus of the above section was acid gas and mercaptan removal. However,
none of the processes discussed address the problem of elemental sulfur removal.
The Engineering Data Book (2004d) briefly describes a polysulfide solution
prepared from elemental sulfur, water, sodium hydroxide, and sodium sulfide
(Na
2
S) for elemental sulfur removal.
2 2
However, COS does not affect the copper strip sulfur test. A properly run copper
strip test includes the addition of water in an attempt to hydrolyze the COS
(Engineering Data Book, 2004a) but may be unreliable. Even when run properly,
the lack of sample vessel conditioning can lead to erroneous results (Eckersley
and Kane, 2004). Tests showed the rate of hydrolysis to be extremely slow in the
copper strip test (McCartney, 2005). (McCartney [2005] notes that the Gas
Processors Association is conducting a study of this problem.) Because of this
problem, a shipment that contains water and COS can pass the test at departure
but fail at delivery. Carbonyl sulfide unfortunately is fairly difficult to eliminate
by conventional methods. Processes for its removal include treatment with either
of the two amines diglycolamine (DGA) or diisopropanolamine (DIPA), or with
adsorption by use of molecular sieve 3A, although the cheaper 4A will work if
the regeneration gas contains no CO
2
. Numerous nonregenerative metallic oxide
processes are also available to remove COS from product streams. Some of these
processes remove the COS directly and others require water to hydrolyze the
COS to H
2
S before it is reacted.
10.3.2 DEHYDRATION
The acceptable water content in NGL streams varies from no free water
present, for process streams, to product specifications of less than 10 ppmw,
for commercial propane and HD-5 propane (Engineering Data Book, 2004a).
As noted in Chapter 6, COS hydrolyzes with water to form CO and H S.
© 2006 by Taylor and Francis Group, LLC
Liquids Processing 231
Figure 10.4, from the Engineering Data Book (2004c), gives the solubility of
shows the solubility of some hydrocarbons in water as a function of temper-
ature. (Yaws et al. [1990] tabulates the solubility of C
4
to C
20
hydrocarbons
in water at 25°C.) However, customers sometimes require very dry liquid
product, such as when it is used as feedstock to a process in which water is
a catalyst poison.
Numerous processes are available for drying liquids, and many of the
FIGURE 10.4 Solubility of water in hydrocarbons as a function of temperature. (Adapted
from Engineering Data Book, 2004c.)
200 180 160 140 120 100 80 60 40
0.005
0.01
0.1
1
Temperature, °F
W
a
t
e
r

s
o
l
u
b
i
l
i
t
y
,

W
t

%

B
e
n
z
e
n
e
T
o
l
u
e
n
e
E
t
h
y
l

b
e
n
z
e
n
e
,

p
-
X
y
l
e
n
e
P
r
o
p
a
n
e
B
u
t
a
n
e
H
e
x
a
n
e
P
e
n
t
a
n
e
0.05
0.5
water in various hydrocarbon liquids as a function of temperature. Figure 10.5
technologies discussed in Chapter 6 for dehydrating gases may be used for liquids.
© 2006 by Taylor and Francis Group, LLC
232 Fundamentals of Natural Gas Processing
Therefore, this section presents only brief descriptions as needed and lists the
pros and cons of each method as they apply to NGL.
10.3.2.1 Adsorption Processes
Section 10.3.1.1 described use of adsorbents in liquid service. When drying, to
have the hydrocarbon flow up the bed to prevent refluxing during regeneration
adequate drying to meet the free water specification. If extremely low water
contents are essential, then the 4A molecular sieve is required.
FIGURE 10.5 Solubility of liquid hydrocarbons in water. (Adapted from Engineering
Data Book, 2004c.)
1 × 10
−6
1 × 10
−5
1 × 10
−4
1 × 10
−3
H
y
d
r
o
c
a
r
b
o
n

s
o
l
u
b
i
l
i
t
y
,

m
o
l
e

f
r
a
c
t
i
o
n
200 180 160 140 120 100 60 80 40
Temperature, °F
p-Xylene
Butane
Toluene
Propane
Benzene
Pentane
Hexane
5 × 10
−6
5 × 10
−5
5 × 10
−4
is particularly advantageous. Any of the adsorbents described in Chapter 6 provide
© 2006 by Taylor and Francis Group, LLC
Liquids Processing 233
10.3.2.2 Desiccant Dehydration
A consumable desiccant such CaCl
2
is viable in remote situations, but disposal
issues make it unattractive compared with the regenerative process in a plant
situation where hydrocarbon liquids are dried. The Engineering Data Book
(2004c) provides a brief description of a two-bed system that generates a solid
waste instead of concentrated brine. The book notes that the process is less used
now because of disposal problems.
10.3.2.3 Gas Stripping
Gas stripping is a simple process for liquid dehydration and is frequently used
offshore to dry condensates before they go to the production platform (Engineer-
ing Data Book, 2004c). The liquid is usually stripped with dry natural gas in a
tower with trays or packing. As shown by Equation 6.1, the process is feasible
because the activity coefficient, g, for water in liquid hydrocarbons is in tens of
thousands, which makes the water volatile.
The process is low cost and space efficient. However, it requires a dry natural
gas stream, which also strips some volatile hydrocarbons at the same time. This
property makes it suitable for only condensates and natural gasoline. The stripping
gas must be compressed and recycled to a dehydration unit or sent to fuel gas.
10.3.2.4 Distillation
Propane and butane mixtures can be dried by use of specially designed columns
(Engineering Data Book 2004c). In these columns, the water goes overhead with
hydrocarbon and is decanted from the overhead reflux drum. The condenser
hydrocarbon returns as reflux. The bottoms product will be bone dry, and even
a side draw three or four trays from the top will meet product specifications. The
Engineering Data Book (2004c) notes that the side-draw liquid must be cooled,
but the process can be cost effective. This factor should make it attractive for
narrow boiling fractions.
10.3.2.5 Absorption
Dehydration of liquid product streams can also be done by use of glycol contact-
ing. These systems are very similar to gas dehydration with glycol, which is
10.3.3 FRACTIONATION
densate stabilizer is included for completeness. As noted previously, the fractionation
train depends upon the current economics and customer requirements. Frequently
the first column takes an ethane−propane mixture (EP mix) overhead instead of pure
ethane. Regardless of whether it is ethane or EP mix, this stream is pipelined to
customers. All of the other products are shipped by tank car, truck, or pipeline.
Table 10.1 provides typical operating conditions for the various columns. The con-
covered in Chapter 6.
© 2006 by Taylor and Francis Group, LLC
234 Fundamentals of Natural Gas Processing
Distillation column design is beyond the scope of this book. However, the
Engineering Data Book (2004b) provides extensive details on column design,
column internals, and reboiler design.
10.4 SAFETY AND ENVIRONMENTAL
CONSIDERATIONS
The major safety and environmental issues involved with treating the hydrocarbon
liquids is the safe handling and disposal of the caustic compounds and solutions,
if used. At present, sulfur specifications are set on the basis of corrosion. Gasoline
and diesel will be required to meet increasingly stringent levels of total sulfur in
fuels. By 2007, the maximum sulfur level in gasoline will be 30 ppm average
(Environmental Protection Agency, 1999). The primary motivation for sulfur reduc-
tion is to reduce NO
X
. NO
X
reduction is accomplished by increasing the perfor-
mance of the automobile’s catalyst to destroy NO
X
, which is poisoned by sulfur.
The push for low sulfur levels in fuel affects the natural gasoline and propane
markets. Natural gasoline may not be accepted as a gasoline blending stock unless
it contains 30 ppm or less of sulfur (McCartney, 2005). The other major market
for the liquid is as a petrochemical feedstock, which often brings a lower sales
price than as a gasoline blending stock. The commonly used propane fuel,
TABLE 10.1
Typical Fractionator Operating Conditions
Operating Pressure,
psig (barg)
Number of
Actual Trays
Reflux
Ratio
a
Reflux

Ratio
b
Tray
Efficiency
(%)
Deethanizer 375 − 450
(26 – 30)
25 − 35 0.9 − 2.0 0.6 − 1.0 60 − 80
Depropanizer 240 − 270
(16 – 19)
30 − 40 1.8 − 3.5 0.9 − 1.1 80 − 90
Debutanizer 70 − 100
(4.8 – 6.2)
25 − 35 1.2 − 1.5 0.8 − 0.9 85 − 95
Butane splitter 80 − 100
(5.5 – 6.9)
60 − 80 6.0 −14 3.0 − 3.5 90 − 100
Condensate
stabilizer
100 − 400
(6.9 – 28)
16 − 24 Top feed Top feed 50 −75
a
Reflux ratio relative to overhead product, mol/mol.
b
Reflux ratio relative to feed, gal/gal.
Source: Adapted from Engineering Data Book (2004b).
© 2006 by Taylor and Francis Group, LLC
Liquids Processing 235
HD-5 propane, could face competition from other low-sulfur fuels. How signif-
icantly this competition will affect the propane market and how gas processors
will respond is unclear at this time.
REFERENCES
Eckersley, N. and Kane, J.A., Designing Customized Desulfurization Systems for the
Treatment of NGL Streams, Proceedings of the Laurance Reid Gas Conditioning
Conference, Norman, OK, 2004, 313.
Engineering Data Book, 12th ed., Sec. 2, Product Specifications, Gas Processors Supply
Association, Tulsa OK, 2004a.
Engineering Data Book, 12th ed., Sec. 19, Fractionation and Absorption, Gas Processors
Supply Association, Tulsa, OK, 2004b.
Engineering Data Book, 12th ed., Sec. 20, Dehydration, Gas Processors Supply Associ-
ation, Tulsa, OK, 2004c.
Engineering Data Book, 12th ed., Sec. 21, Hydrocarbon Treating, Gas Processors Supply
Association, Tulsa, OK, 2004d.
Environmental Protection Agency, EPA’s Program for Cleaner Vehicles and Cleaner Gasoline,
2005.
Kuchinski, J. private communication, 2005.
Langston, L.V., The Lease Pumper’s Handbook, 1st ed., The Oklahoma Commission on
McCartney, D.G., private communication, 2005.
Mick, M.B., Treat propane for COS removal, Hydrocarbon Process., 55 (7) 137, 1976.
Nielsen, R.B., Principles of LPG Amine Treater Design, Proceedings of the Laurance Reid
Gas Conditioning Conference, Norman, OK, 1995, 229.
Richman, P., private communication, 2005.
Sargent, A. and Seagraves, J., LPG Contactor Design and Practical Troubleshooting Tech-
niques, Proceedings of the Laurance Reid Gas Conditioning Conference, Norman,
OK, 2003, 125.
Sambrook, R.M., Dillingham, B., and Prugh, J., A Liquid Propane Desulfurization Study,
Proceedings of the Laurance Reid Gas Conditioning Conference, Norman, OK,
1997, 158.
Verachtert, T.A., Cassidy, R.T., and Holmes, E.S., The UOP Merox Process for NGL Treat-
ing, Proceedings of the Laurance Reid Gas Conditioning Conference, Norman, OK,
1990, 267.
Veroba, R. and Stewart, E., Fundamentals of Gas Sweetening, Proceedings of the Laurance
Reid Gas Conditioning Conference, Norman, OK, 2003, 1.
Webber, W.W., Petty, L.E., Ray, B.D., and Story, J.G., Whitney Canyon Project, Proceed-
ings of the Sixty-Third Annual Convention if Gas Processors Association, Tulsa,
OK, 1984, 105.
Yaws, C. L., Yang, H-C., Hopper, J.R., and Hansen, K.C., Hydrocarbons: Water solubility
data, Chemical Eng., 97 (4) 1990, 177.
Marginally Producing Oil and Gas Wells, Norman, OK, 2003, http://www.mar-
ginalwells.com/MWC/MWC/Searchable%20Text2.pdf, Retrieved May 2005.
EPA420-F-99-051, 1999, www.epa.gov/tier2/frm/f99051.pdf, Retrieved June
© 2006 by Taylor and Francis Group, LLC
237
11
Sulfur Recovery
11.1 INTRODUCTION
Leppin (2001) points out that approximately 25% of the natural gas being brought
into production from new sources requires H
2
S removal and disposal. Conse-
Field Liquids Removal
Gas & Liquids from Wells
Inlet
Compression
Sulfur
Recovery
Gas Treating
Dehydration
Nitrogen
Rejection
Helium
Recovery
Outlet
Compression
Elemental
Sulfur
Field Dehydration
Field Acid Gas Removal
Field Compression
Sales Gas
Water &
Solids
Natural
Gas Liquids
Natural
Gasoline
Raw Helium
CO
2
Liquefaction
Liquefied
Natural Gas
Inlet Receiving
Hydrocarbon
Recovery
Liquids Processing
quently, sulfur removal processes discussed in Chapter 5 will play an increasingly
© 2006 by Taylor and Francis Group, LLC
238 Fundamentals of Natural Gas Processing
larger role in future gas processing. Currently only two methods are available for
dealing with large quantities of H
2
S:
• Disposal of the gas by injection into underground formations
• Conversion of the H
2
S into a usable product, elemental sulfur
Where environmental regulations permit and production of elemental sulfur
is not economically attractive, injection wells provide a safe means for H
2
S
disposal. Jones et al. (2004) discuss design and operation of a facility that enriches
an H
2
S stream before injection into a disposal well. Oberding et al. (2004) provide
additional details on the project.
However, more commonly, H
2
S is converted into elemental sulfur, much of
which goes into sulfuric acid production. As late as 1950, over half of the world’s
sulfur supply came from “voluntary producers,” that is, companies whose principal
purpose was to produce elemental sulfur. Now, these producers furnish less than
5% of the world’s supply and “involuntary producers,” primarily petroleum refin-
eries and natural gas plants, are the major source of the element (Hyne, 2005).
The most common method of converting H
2
S into elemental sulfur, is the Claus
process or one of its modifications. We discuss two modified Claus configurations:
straight through and split flow (Engineering Data Book, 2004).
The straight-through process is the preferred and simplest, and it can process
feed streams that contain more than 55 mol% H
2
S; with air or acid gas preheat,
it can process 30 to 55 mol% H
2
S in the feed. The split-flow configuration can
process feeds that contain 5 to 30 mol% H
2
S. Lower concentration feeds require
other variations (Engineering Data Book, 2004) and are not discussed here.
The straight-through process provides the highest sulfur-recovery efficiency.
Unfortunately, the exit stream from Claus plants usually cannot meet envi-
ronmental emission requirements, and, consequently, a tail gas cleanup unit
(TGCU) is often employed to eliminate the last of the sulfur compounds. Accord-
ing to available literature, the most commonly used processes are Shell Claus
Offgas Treating (SCOT), SUPERCLAUS, and cold-bed adsorption (CBA). This
discussion will be limited to these processes.
The most comprehensive information on sulfur recovery processes is avail-
able in the Conference Fundamentals Manual of the Laurance Reid Gas Con-
ditioning Conference (2003), Maddox and Morgan (1998), Engineering Data
Book (2004), Kohl and Nielsen (1997), Echterhoff (1991), and Leppin (2001).
Strickland et al. (2001) provide a thorough discussion of TGCU processes,
along with cost comparisons.
This chapter discusses the properties of elemental sulfur and then describes
Claus and tail gas cleanup processes. It then briefly discusses sulfur storage.
11.2 PROPERTIES OF SULFUR
The thermophysical properties of sulfur are unusual. For a complete understand-
ing of the Claus conversion process, a brief discussion of the relevant properties
is necessary.
© 2006 by Taylor and Francis Group, LLC
Sulfur Recovery 239
Sulfur vapor exists as S
x
, where x can have values from 1 through 8. Figure 11.1
shows the distribution of sulfur-vapor species as a function of temperature. Note
that at lower temperatures, S
8
dominates, but as the temperature rises S
8
converts to
S
6
, and finally to S
2
. The sequence is not unexpected, because increased temperature
means increased energy for the molecules, which leads to the breakup of the clusters.
The formation of sulfur clusters has a very pronounced effect on the physical
properties that have a significant effect on processing operations, notably viscosity
(fluid flow) and heat capacity (heat transfer). An excellent summary of liquid and
vapor properties is available in the Engineering Data Book (2004).
11.3 SULFUR RECOVERY PROCESSES
11.3.1 CLAUS PROCESS
All Claus units involve an initial combustion step in a furnace. The combustion
products then pass through a series of catalytic converters, each of which produces
elemental sulfur. We briefly discuss the basic chemistry and then describe process
details.
11.3.1.1 Basic Chemistry
The Claus process consists of the vapor-phase oxidation of hydrogen sulfide to
form water and elemental sulfur, according to the overall reaction:
3 H
2
S + 3/2 O
2
→ 3 H
2
O + (3/x) S
x
. (11.1)
FIGURE 11.1 Sulfur-vapor species as a function of temperature. (Adapted from Engi-
neering Data Book, 2004.)
0.2
0.0
1200 1400 600 400 200
M
o
l

f
r
a
c
t
i
o
n

o
f

s
p
e
c
i
e
s
S
2
S
6
S
8
800 1000
0.4
0.6
0.8
1.0
Temperature, °F
© 2006 by Taylor and Francis Group, LLC
240 Fundamentals of Natural Gas Processing
The above overall reaction does not represent the reaction mechanism or
show intermediate steps. In practice, the reaction is carried out in two steps:
H
2
S + 3/2 O
2
↔ H
2
O + SO
2
(11.2)
2 H
2
S + SO
2
↔ 2 H
2
O + (3/x) S
x
. (11.3)
The first reaction is a highly exothermic combustion reaction, whereas the
second is a more weakly exothermic reaction promoted by a catalyst to reach
equilibrium. Figure 11.2 shows the equilibrium conversion obtained for H
2
S into
elemental sulfur by the Claus reaction. Kohl and Nielsen (1997) state that the
unusual shape of the equilibrium curve is caused by the existence of different
sulfur species at different reaction temperatures. They point out that at a sulfur
partial pressure of 0.7 psia (0.05 bar) and temperatures below 700°F (370°C),
the vapor is mostly S
6
and S
8
, but at the same partial pressure and temperatures
over approximately 1,000°F (540°C), S
2
predominates. This shift in species
causes the equilibrium constant in the reaction to shift from a downward slope
to an upward slope, as shown in Figure 11.2. This behavior has a significant
effect on the operation of the Claus process.
The melting point of amorphous sulfur is 248°F (120°C), and its normal
boiling point is 832°F (445°C). Figure 11.2 shows that the maximum conversion
to sulfur by reaction 11.1 is obtained at temperatures near the melting point of
sulfur, but to maintain sulfur in the vapor state, relatively high temperatures are
required. Consequently, if the catalytic converters are to operate under conditions
in which the sulfur does not condense on the catalyst, they cannot operate at
optimum equilibrium conversion. This is the reason for having a series of con-
verters, with the sulfur product withdrawn from the reacting mixture between
converters. Withdrawing the sulfur product causes the reaction 11.3 to shift to
the right, which results in more sulfur product.
FIGURE 11.2 Equilibrium conversion of hydrogen sulfide to sulfur. (Adapted from Engi-
neering Data Book, 2004.)
2400 2000 1600 1200 800 400 0
P
e
r
c
e
n
t

c
o
n
v
e
r
s
i
o
n

o
f


H
2
S

t
o

s
u
l
f
u
r

40
60
80
100
2800
Temperature, °F
© 2006 by Taylor and Francis Group, LLC
Sulfur Recovery 241
11.3.1.2 Process
Figure 11.3 and Figure 11.4 show simplified flow diagrams for the two common
configurations, straight through and split flow, respectively. In the straight-through
configuration, the first reaction takes place in a combustion furnace operating
near ambient pressure (3 to 8 psig; 0.2 to 0.6 barg). The air flow rate is adjusted
to react with one third of the H
2
S, along with any other combustibles, such as
hydrocarbons and mercaptans. The H
2
S reaction is exothermic (637 Btu/scf, or
24,000 kJ/m
3
,

at 25°C at 1 atm) and is used to produce steam in a waste-heat
boiler. Both reactions take place in the furnace−boiler combination, and the gases
exit the waste-heat boiler in the range of 500 to 650°F (260 to 343°C) (Parnell,
1985), which is above the sulfur dew point, so no sulfur condenses in the boiler.
FIGURE 11.3 Straight-through Claus unit. (Adapted from Echterhoff, 1991.)
FIGURE 11.4 Split-flow Claus unit. Typically two thirds of feed bypasses the reaction
furnace. (Adapted from Echterhoff, 1991.)
Tail gas
condense
r
LP
steam
LP
steam
Air
Acid gas
HP
Steam
Condenser Condenser Condenser
Claus #1 Claus #2
LP
Steam
Boiler feed
water (BFW)
Condenser
Claus #3
110°F
8 psig
600°F
375°F
450°F
590°F
350°F
420°F
453°F 300°F
400°F
407°F
270°F
Sulfur
BFW
Sulfur
BFW
Sulfur
BFW
Sulfur
BFW
Heated
BFW
Reaction
furnace
Waste heat
boiler
Gas
reheat
Gas
reheat
Gas
reheat
Tail gas
condense
r
LP
steam
LP
steam
Air
Acid gas
HP
steam
Condenser Condenser Condenser
Claus #1 Claus #2
LP
steam
Boiler feed
water (BFW)
Condenser
Claus #3
Sulfur
BFW
Sulfur
BFW
Sulfur
BFW
Sulfur
BFW
Heated
BFW
Gas
reheat
Gas
reheat
© 2006 by Taylor and Francis Group, LLC
242 Fundamentals of Natural Gas Processing
The combustion furnace−boiler is followed by several catalytic reactors in
which only the second reaction takes place because all the O
2
has been consumed
in the furnace. Each catalytic reactor is followed by a condenser to remove the
sulfur formed. The gas is cooled to 300 to 400°F (149 to 204°C) in the condenser
to remove elemental sulfur. The condenser generally achieves cooling by heat
exchange with water to produce low-pressure steam. Vapor that leaves the con-
denser is at the sulfur dew point, so the gas is reheated before passing to the next
converter to prevent sulfur deposition on the catalyst.
A combustion-furnace flame temperature of 1700°F (927°C) should be main-
tained (Kohl and Nielsen, 1997) because the flame is not stable below this value.
The straight-through configuration cannot be used at H
2
S concentrations below
55%, because the feed gas heating value is too low. Concentrations as low as
40% are acceptable (Parnell, 1985), if the air or acid gas is preheated.
For H
2
S concentrations in the range 25 to 40%, the split flow configuration
more of the feed goes to the furnace and the remainder joins the furnace exit gas
before entering the first catalytic converter. When two thirds of the feed is
bypassed, the combustion air is adjusted to oxidize all the H
2
S to SO
2
, and,
consequently, the necessary flame temperature can be maintained (Kohl and
Nielsen, 1997). The split-flow process has two constraints (Kohl and Nielsen, 1997):
1. Sufficient gas must be bypassed so that the flame temperature is greater
than approximately 1,700°F (927°C).
2. Maximum bypass is two thirds because one third of the H
2
S must be
reacted to form SO
2
. (Note the stoichiometry in equations 11.2 and 11.3.)
If air preheating is used with the split-flow configuration, gases with as little
as 7% H
2
S can be processed (Parnell, 1985).
Generally, the sulfur recovery in the conventional plants discussed above
varies from 90 to 96% for two catalytic converters. It increases to 95 to 98% for
three catalytic converters (Lagas et al., 1989).
11.3.2 CLAUS TAIL GAS CLEANUP
Before sulfur emissions restrictions were imposed, the offgas from the Claus unit
was flared to convert the remaining H
2
S to SO
2
. Now, environmental agencies
demand higher sulfur recovery than can be achieved with a standard Claus unit,
and additional treating of the Claus tail gas is needed. The tail gas cleanup entails
either an add-on at the end of the Claus unit or a modification of the Claus unit
itself. The processes for this final sulfur removal are generally divided into three
categories (Kohl and Nielsen, 1997):
• Direct oxidation of H
2
S to sulfur
• Sub-dew point Claus processes
• SO
2
reduction and recovery of H
2
S
(Figure 11.4) can be utilized. In this scheme, the feed is split, and one third or
© 2006 by Taylor and Francis Group, LLC
Sulfur Recovery 243
d’Haêne (2003) provides a recent overview of tail gas cleanup processes.
This section provides an example process in each category.
11.3.2.1 Direct Oxidation of H
2
S to Sulfur
SUPERCLAUS is an example of selective oxidation for final sulfur removal. The
process as described by Lagas et al. (1988, 1989, 1994) involves a slightly modified
two-stage Claus unit followed by a third-stage catalytic reactor to oxidize the
2
Reactors 1 and 2 use the standard Claus catalyst, whereas the third reactor
contains the selective oxidation catalyst. The Claus unit itself is operated with
a deficiency of air so that the gas that exits the second reactor contains 0.8 to
3 vol% H
2
S. Sufficient air is added to this exit gas to keep the oxygen level in
the 0.5 to 2 vol% range. The mixture then goes to the third reactor, where the
following catalytic reaction occurs:
2H
2
S + O
2
→ 2S + 2H
2
O (11.4)
The selective oxidation catalyst in the third reactor does not promote the reaction
2H
2
S + 3O
2
↔ 2SO
2
+ 2H
2
O (11.5)
or the reverse reaction of sulfur with H
2
O
3S + 2H
2
O ↔ 2H
2
S + SO
2
(11.6)
and, consequently, a total recovery rate of 99% or higher can be reached. Lagas,
et al. (1988) describes a modification of SUPERCLAUS that can reach a sulfur
recovery rate of 99.5%.
11.3.2.2 Sub-Dew Point Claus Processes
This category has several processes, but the only one considered here is the cold-
bed adsorption (CBA) process because it is the most widely used sub-dew point
(Goddin et al., 1974). The front end of the unit is a Claus reactor, and the sub−
dew point process takes place in the final two catalytic converters, CBA #1 and
CBA #2. One converter is in service while the other is being regenerated. After
the gas leaves the final condenser, it is not reheated but is sent to the third converter
(CBA #2), which is operated at a temperature well below the sulfur dew point.
This flow results in a better equilibrium conversion but deposits sulfur on the
catalyst and causes a gradual loss of activity. During the period that CBA #2
is in service, CBA #1 is being heated and regenerated with a slip stream from
Claus #1. After regeneration is complete, the reactor can be cooled with a slip
stream from the final condenser. When the catalyst in CBA #2 is exhausted, valves
are switched, and CBA #1 becomes the adsorbing reactor while CBA #2 under-
goes regeneration. Total sulfur recoveries greater than 99% can be obtained with
CBA processes (Clark et al., 2002)
remaining H S to elemental sulfur. Figure 11.5 shows the overall process.
process (Ortloff, undated). Figure 11.6 shows one of its many possible variations
S
u
l
f
u
r

R
e
c
o
v
e
r
y
2
4
4
FIGURE 11.5 SUPERCLAUS unit. (Adapted from d’Haene, 2003.)
Sulfur
Air
Acid
gas
BFW BFW
HP
steam
LP
steam
LP
steam
LP
steam
LP
steam
Claus #1
Claus #2
BFW
BFW
BFW
qc
qc
Sulfur
fc
fc
Control H
2
S to 0.8–1.5 %
Control O
2
to
0.5 to 2.0%
Sulfur Sulfur
Selective
oxidation reactor
Incinerator
To
stack
Condenser Condenser Condenser
Condenser
© 2006 by Taylor and Francis Group, LLC
© 2006 by Taylor and Francis Group, LLC
Sulfur Recovery 245
11.3.2.3 Reduction of SO
2
and Recovery of H
2
S
The SCOT process is an example of the process that reduces the SO
2
in the Claus
plant offgas back to H
2
S. It then uses amine treating to remove the H
2
S, which
is recycled back to the Claus plant for conversion to elemental sulfur. The process
can produce an exit gas that contains 10 to 400 ppmv of total sulfur (Kohl and
Nielsen, 1997), while increasing total sulfur recovery to 99.7% or higher. Figure 11.7
shows a simplified process flow diagram of the SCOT process.
The following description of the process has been adapted from Echterhoff
(1991). The feed, offgas from the Claus unit, is heated to 575°F (302°C) in an inline
FIGURE 11.6 Cold-bed adsorption unit. Dashed line denotes regeneration stream.
(Adapted from Goddin et al., 1974.)
FIGURE 11.7 Shell Claus Offgas Treating (SCOT) unit. (Adapted from Echterhoff, 1991.)
condense
r
Air
Acid gas
HP
steam
Condenser Condenser Condenser
Claus #1 Claus #2
LP
steam
LP
steam
LP
steam
BFW
Sulfur Sulfur Sulfur
BFW BFW BFW
Gas
reheat
Gas
reheat
Gas
reheat
Tail gas
CBA #2
adsorber
Sulfur
Condenser
CBA #1
adsorber
Claus tail gas
Reducing gas
Air
Fuel
Catalytic reactor
Low pressure
steam reboiler
H
2
O
To stack
H
2
S recycle
Low pressure
steam
Q
u
e
n
c
h

t
o
w
e
r
A
b
s
o
r
b
e
r
R
e
g
e
n
e
r
a
t
o
r
Furnace
© 2006 by Taylor and Francis Group, LLC
246 Fundamentals of Natural Gas Processing
burner, along with a reducing gas, H
2
or a CO and H
2
mixture. The reducing gas is
supplied either from an outside source or generated by partial oxidation in an inline
burner. The mixture then flows to the SCOT catalytic reactor (cobalt−molybdenum
on alumina), where the sulfur compounds, including SO
2
, CS
2
, and COS are reduced
to H
2
S and water. The gas that leaves the reactor goes to a waste-heat exchanger,
where it is cooled to about 320°F (160°C) and produces low-pressure steam.
The gas from the waste-heat exchanger then flows through a quench tower,
where it is cooled to approximately 100°F (38°C) by externally cooled recycle water
in countercurrent flow. The water from the tower is condensed, and the excess water
is sent to a sour water stripper. Gas from the quench tower then contacts an aqueous
amine solution in the absorption column. The amine is generally methyldiethanola-
mine (MDEA) or diisopropylamine (DIPA) to absorb H
2
S while slipping CO
2
. The
gas that exits the top of the absorber contains very little H
2
S (10 to 400 ppm) and
is sent to an incinerator. The rich amine that leaves the bottom of the absorber flows
to the regenerator, where heat is applied to strip the H
2
S from the amine solution.
The overhead from the regenerator is cooled to condense the water, and the H
2
S is
recycled to the Claus unit. Lean amine is cooled and returned to the absorber.
11.4 SULFUR STORAGE
Sulfur from the Claus unit is withdrawn as a liquid and is generally stored and
transported in the molten state. A number of potential problems are associated
with sulfur storage, including release of H
2
S dissolved in the molten sulfur and
the possibility of sulfur fires, which will produce highly toxic SO
2
. A brief
summary, adapted from Johnson and Hatcher (2003a), is given below:
• Sulfur fires. Uncommon, but they can produce large amounts of SO
2
.
• H
2
S. Any H
2
S dissolved in the molten sulfur from the condensers may
be a significant hazard if appropriate degassing techniques are not used.
• Corrosion. A wet sulfidic atmosphere can lead to severe corrosion of
carbon steel.
• SO
2
. Highly toxic and it forms highly corrosive sulfurous acid in the
presence of water.
• Static discharge. Because of the excellent insulating properties of mol-
ten sulphur, static discharge may occur under certain conditions and
lead to possible fires or explosions.
A comprehensive discussion of sulfur storage and degassing is beyond the scope
of this book, but Johnson and Hatcher (2003a, 2003b) give an excellent presentation.
11.5 SAFETY AND ENVIRONMENTAL CONSIDERATIONS
marize the more important physical properties of molten sulfur, hydrogen sulfide,
and sulfur dioxide. The tables also address safety and environmental concerns.
Table 11.1 through Table 11.3, adapted from Johnson and Hatcher (2003a), sum-
© 2006 by Taylor and Francis Group, LLC
Sulfur Recovery 247
TABLE 11.1
Important Properties, Safety, and Environmental Concerns
of Molten Sulfur
Physical Property
or Characteristic Hazards
Melts at 240°F ( 116°C)
a
Possible severe burns
Boils at 851°F (455°C)
a
Exhibits measurable vapor pressure at molten
sulfur handling process conditions
Flash point 334 − 369°F (168 − 187°C) Not much higher that normal storage temperatures
Autoignites at 478 − 511°F
(248 − 266°C)
Sulfur fires can easily occur in the presence of oxygen
Viscosity transition at 318°F (159°C) Increasingly difficult to pump above this temperature
Magnitude is dependent upon H
2
S content
of the molten sulphur
Highly flammable in air Sulfur fires
Solid dust cloud autoignites at 375°F
(191°C)
Important variable in solid-forming operations
Deflagration possible
Excellent insulator Static discharge at 11% or higher oxygen atmosphere
Toxicity Highly toxic at low concentrations
a
For S
8
,

Georgia Gulf Sulfur (2000) gives a melting point of 230 to 246°F (110 to119°C)
and
a
Boiling point of 832°F (444°C).
Source: Johnson and Hatcher (2003a).
TABLE 11.2
Important Properties, Safety, and Environmental
Concerns of H
2
S.
Physical Property or Characteristic Hazards
Highly flammable in air between
approximately 3.4 and 45%
at process conditions
Deflagration possible
Exposed carbon steel forms iron
sulfide corrosion product under wet,
reducing conditions
Excessive corrosion in carbon steel piping
in some cases
Iron sulfides are highly pyrophoric
Source: Johnson and Hatcher (2003a).
© 2006 by Taylor and Francis Group, LLC
248 Fundamentals of Natural Gas Processing
REFERENCES
Clark, P., et al., Enhancing the Performance of the CBA Process by Optimizing Catalyst
Macroporosity, Proceedings of the Laurance Reid Gas Conditioning Conference,
Norman, OK, 2002, 79.
Conference Fundamentals Manual, Proceedings of the Laurance Reid Gas Conditioning
Conference, Norman, OK, 2003.
d’Haêne, P.E., Tail Gas Treating, Proceedings of the Laurance Reid Gas Conditioning
Conference, Norman, OK, 2003, 89.
Echterhoff, L.W., State of the Art of Natural Gas Processing Technologies, Gas Research
Institute Report, GRI-91/0094, 1991.
Engineering Data Book, 12th ed., Sec. 22, Sulfur Recovery, Gas Processors Supply Asso-
ciation, Tulsa, OK, 2004.
Retrieved September 2005.
Goddin, C.S., Hunt, E.B. and Palm, J.W., CBA process ups Claus recovery, Hydrocarbon
Process., 53 (10) 122, 1974.
Hyne, J.B., The sulfur bubble, Hydrocarbon Eng., 10 (4) 23, 2005.
Johnson, J.E. and Hatcher, N.A., Hazards of Molten Sulfur Storage and Handling, Pro-
ceedings of the Laurance Reid Gas Conditioning Conference, Norman, OK, 2003a,
109.
Johnson, J.E. and Hatcher, N.A., A comparison of established sulfur degassing technolo-
gies, Proceedings of the Laurance Reid Gas Conditioning Conference, Norman,
OK, 2003b, 131.
Jones, S.G., Rosa, D.R., and Johnson, J.E., Design, Cost and Operation of an Acid Gas
Enrichment and Injection Facility, Proceedings of the Laurance Reid Gas Condi-
tioning Conference, Norman, OK, 2004, 81.
Kohl, A. L. and Nielsen, R.B., Gas Purification, Gulf Publishing, Houston, TX, 1997.
Lagas, J.A., Borsboom, J., and Berben, P.H., The SUPERCLAUS Process, Proceedings
of the Laurance Reid Gas Conditioning Conference, Norman, OK, 1988, 41.
Lagas, J.A., Borsboom, J., and Heijkoop, G., Claus process gets extra boost, Hydrocarbon
Process., 68 (4) 40, 1989.
Lagas, J.A., Borsboom, J., and Goar, B.G., SUPERCLAUS, Five Years of Operating
Experience, Proceedings of the Laurance Reid Gas Conditioning Conference,
Norman, OK, 1994, 192.
Leppin, D., Large-scale sulfur recovery, GasTIPS, 7, 26, 2001.
TABLE 11.3
Important Properties, Safety, and Environmental
Concerns of SO
2

Toxicity Highly toxic at low concentrations
Corrosive when wet Forms sulfurous acid in presence of water at
low temperatures
Source: Johnson and Hatcher (2003a).
Georgia Gulf Sulfur Corporation, www.georgiagulfsulfur.com/properties.htm, 2000,
© 2006 by Taylor and Francis Group, LLC
Sulfur Recovery 249
Maddox, R.N. and Morgan, D. J., Gas Conditioning and Processing, Vol. 4, Gas Treating
and Sulfur Recovery, Campbell Petroleum Series, Norman, OK, 1998.
Oberding, W., et al., Lessons Learned and Technology Improvements at the Lost Cabin
Gas Plant, Proceedings of the Laurance Reid Gas Conditioning Conference,
Norman, OK, 2004, 273.
Parnell, D., Look at Claus unit design, Hydrocarbon Process., 64 (9) 114, 1985.
Strickland, J.F., et al., Relative capabilities and costs of tail gas clean-up processes,
GasTIPS, 7 9, 2001.
Ortloff Engineers, Ltd, www.Ortloff.com, Retrieved July 2005.
© 2006 by Taylor and Francis Group, LLC
251
12
Transportation
and Storage
12.1 INTRODUCTION
Whereas raw natural gas is usually located relatively close to gas plants, the
processed natural gas and natural gas liquid products must be transported to
the end-user, industrial or residential. Except for the less common situation in
which the gas plant is dedicated to a single commercial customer, product
storage is required. In the case of natural gas, transportation and storage prob-
lems are more difficult than with other common forms of energy, such as coal
and oil, because the energy density of natural gas is so low at ambient temper-
atures and pressures. For an equal volume of these three fuels at ambient
temperature and pressure, the energy content of gasoline is approximately
1,000 times greater than that of natural gas. The same volume of coal (anthra-
cite, bulk density 50 to 58 lb/ft
3
[800 to 940 kg/m
3
]), with a heating value
13,500 Btu/lb (31 MJ/kg), contains 700 times more energy as a fuel. This large
difference in energy density compared with natural gas highlights two major
problems with natural gas. First, a relatively high pressure is required to increase
the gas density and raise the energy content per unit volume so that the gas
can be transported economically by pipeline (common pressures are approxi-
mately 800 to 1,500 psig [60 to 100 barg]). Second, large quantities of natural
gas cannot be stored in relatively simple and inexpensive aboveground facilities
similar to those used for liquid-petroleum products. Note that for methane, the
primary component of natural gas, the critical temperature is −118°F (−83°C),
and, consequently, no amount of pressure can convert methane into a liquid at
60°F (15°C).
of hydrocarbons that can be removed and liquefied (NGL) to produce the dry
natural gas distributed to industrial and residential customers. (Dry in this context
means that the NGL content is sufficiently low so that liquid will not drop out
at pipeline temperatures and pressures.) Although more volatile, NGL is trans-
ported and stored in much the same way as crude oil and refined petroleum
products.
Natural gas can be liquefied at cryogenic temperatures as liquefied natural
transporting and storing LNG.
As discussed in Chapter 1, many natural gases contain significant quantities
gas (LNG). The following chapter discusses the unique issues associated with
© 2006 by Taylor and Francis Group, LLC
252 Fundamentals of Natural Gas Processing
12.2 GAS
12.2.1 TRANSPORTATION
Depending on the gas plant’s function, any of several possible gas streams are
transported from the plant to customers. Producers ship less common gas products
such as hydrogen sulfide, carbon dioxide, nitrogen, and raw helium to customers
via dedicated pipelines. These lines tend to be short because the customer usually
is located near the gas plant. Although some plants have a single dedicated
customer for the sales gas, most gas goes to an extensive pipeline network
dedicated to natural gas service. Because product quality is so uniform within
each country, many different gas processors utilize the same trunk pipelines, and
many gas plants are connected to multiple pipelines.
The United States in 2004 had over 212,000 miles of interstate gas pipelines
(Energy Information Administration 2005a), serviced by more than 1,200 com-
pressor stations, and capable of transporting over 32 Tcf (910 Bm
3
) of natural
gas per year (Smith et al., 2005).
The average increase in pipeline capacity (excluding gathering system lines)
between 1998 and 2004 was close to 9 Bcfd (0.25 Bm
3
/d), but this capacity is
expected to increase even further over the next 3 years (Energy Information
Administration, 2005a). This increase corresponds to an average of about 2,300
miles (3700 km) of pipelines added each year during the same period. Of these
new pipelines, six major pipeline systems were put into operation in the Gulf of
Mexico in 2004. This capacity accounted for 1.8 Bcfd (0.05 Bm
3
/d), which was
23% of the total increase (Energy Information Administration, 2005a).
Figure 12.1 shows the distribution of line diameters scheduled for construction
in 2006 in both the United States and the world; however, information on operating
pressures for the lines was not available. The smaller diameter lines are normally
the wellhead to the gas processing plant. The larger lines are generally long distance
transmission lines designed to bring sales gas to customers.
The larger and longer lines are normally built with 30-inch or more (75 cm)
diameters. For example, one proposal for a gas pipeline from Alaska to the
lower 48 states calls for a 48-inch (120-cm) line that operates at 2,500 psi
FIGURE 12.1 Distribution of line diameters in the United States and in the World Sched-
uled for Construction in 2006 (True, 2004).
World pipeline diameters U.S. pipeline diameters
12–20
inch, 32.0%
12–20
inch, 25.3%
22–30
inch, 29.5%
22–30
inch, 40.3%
4–10 inch,
2.9%
4–10 inch,
1.7%
30+inch,
35.6%
30+inch,
32.7%
used in the gathering systems, discussed in Chapter 3, which bring the gas from
© 2006 by Taylor and Francis Group, LLC
Transportation and Storage 253
(170 bar), and the West to East pipeline in China will have a mainline of 2,360
miles (3,800 km), with a diameter of 40 inches (100 cm), and 183 miles (294 km)
of lateral lines, with diameters of 32 inches (813 mm), 20 inches (50 cm), and
16 inches (40 cm) (Gray et al., 2003). Major interstate pipelines are usually 24
to 36 inches (60 to 90 cm) in diameter, and the lateral lines that connect to the
mainline are typically 6 to 16 inches (15 to 40 cm). More detailed information
on pipeline transportation is available in the book by Manning and Thompson
(1991).
Thirty gas pipeline compressor stations were built in the United States in
2003 (True and Stell, 2004), and 13 were built in 2004 (Smith et al., 2005).
Capital costs averaged slightly over $1,200/hp, but the average can be misleading
because costs ranged from $748/hp to $3,436/hp. The large variation is expected,
as the data do not consider important cost factors, such as location and type of
compressor used. These cost figures include materials, labor, land, and miscella-
neous items. Figure 12.2 shows the relationship between total cost and station
size. A definite, expected relation exists between cost and horsepower. An approx-
imate cost for new gas pipeline compressor stations can be determined from the
following equation:
Cost (10
6
US $) = 2.6 + 1.1 (hp/1,000).
The cost for a compressor station on gathering systems may be as much
as 50% lower (McCartney, 2005). Note that these costs are in 2004 dollars.
FIGURE 12.2 Total cost of gas pipeline compressor stations (millions of US dollars) as
a function of compression horsepower (True and Stell, 2004 (open symbol); Smith et al.,
2005 (closed symbol)). See text for details.
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0
100.0
0 20 40 60 80 100
Compressor power, 1000 Hp
C
o
m
p
r
e
s
s
o
r

c
o
s
t
,

1
0
6

$
Chapter 15 discusses how to correct for inflation.
© 2006 by Taylor and Francis Group, LLC
254 Fundamentals of Natural Gas Processing
12.2.2 MARKET CENTERS
Market centers, or hubs, are usually located near the intersection of several major
pipelines. They are an outgrowth of market restructuring in the 1980s and the require-
ment that interstate gas companies transport gas but not buy or sell the gas. They
exist in the United States and in Canada. According to the Energy Information
Administration (2003) “The defining characteristics of a natural gas market center
are that it provides customers (shippers and gas marketers primarily) with
receipt/delivery access to two or more pipelines systems, provides transportation
between these points, and offers administrative services that facilitate that movement
and/or transfer of gas ownership.” In 2003, 37 operational hubs existed in the United
States and Canada. Market center services vary widely between facilities, and no two
hubs provided the services through the Internet; the others used fax-phone. The
business is highly competitive, and like any business has had closings and mergers.
The best known (but not the largest) of the United States and Canadian
market centers is undoubtedly the Henry Hub, located in Louisiana and serving
an average of 180 large commercial customers each day through 14 intercon-
necting pipeline systems. The hub can handle over 1.8 Bscfd (0.05 Bm
3
) of
gas, has two salt dome storage sites with a working capacity of 15 Bcf (0.42
Bm
3
), and also serves as the delivery point for New York Mercantile Exchange
(NYMEX) natural gas futures contracts. Market centers are valuable in facil-
itating the management of transportation and storage of natural gas in the
United States, and their numbers and importance will undoubtedly increase in
the future. Additional information can be obtained from the Energy Information
Administration (2003) and Beckman and Determeyer (1997).
12.2.3 STORAGE
12.2.3.1 Introduction
Before 1992, the natural gas market was regulated, and natural gas storage had
two main purposes:
• To provide baseload storage to meet seasonal, weather-sensitive,
demands above normal pipeline delivery capability
• To provide peak (or peak shaving) storage to smooth out the demand
curve
After deregulation, storage facilities were open for additional uses, such as
storing gas when prices were low and selling gas when prices were higher. The
storage can be either baseload or peak shaving. Baseload storage provides a
volume of working gas for a long-term steady withdrawal of the gas (typically
3 to 5 months), followed by a long-term injection period (typically 5 to 7 months).
Peak shaving facilities, on the other hand, are designed to provide large quantities
hubs are identical, but Table 12.1 lists typical services provided. In 2003, all but seven
of gas over a relatively short time frame (hours or days). Figure 12.3 clearly
© 2006 by Taylor and Francis Group, LLC
Transportation and Storage 255
illustrates the necessity for both types of storage by showing how the demand
for natural gas in the United States cycled over a 3-year period. Considering the
very large peaks and valleys, facilities capable of both meeting the long-term
“average” demand and possessing storage to handle the peak seasonal demand
are obviously necessary. Note that the spike in the summer months demand results
from electric utilities that use gas turbines (the same as those used for gas
compression) to meet peak power demands.
12.2.3.2 Facilities
Basically, two types of storage facilities exists for natural gas: (1) relatively small
capacity (to 15 MMscf [400,000 Sm
3
]) aboveground, floating-roof gas holders that
TABLE 12.1
Services Arranged and Coordinated in Market Hub Centers
Transportation/wheeling—Transfer of gas from one interconnected pipeline to another through a
header (hub), by displacement (including exchanges), or by physical transfer over the
transmission of a market-center pipeline.
Parking—A short-term transaction in which the market center holds the shippers gas for redelivery
at a later date. Often uses storage facilities, but may also use displacement or variations in
linepack.
Linepack—Refers to the gas volume contained in a pipeline segment. Linepack can be increased
beyond the pipeline’s certificated capacity temporarily and, within tolerances, by increased
compression.
Loaning —A short-term advance of gas to a shipper by a market center that is repaid in kind by
the shipper a short time later. Also referred to as advancing, drafting, reverse parking, and
imbalance resolution.
Storage—Storage that is longer than parking, such as seasonal storage. Injection and withdrawal
operations may be separately charged.
Peaking —Short-term (usually less than a day and perhaps hourly) sales of gas to meet
unanticipated increases in demand or shortages of gas experienced by the buyer.
Balancing—A short-term interruptible arrangement to cover a temporary imbalance situation. The
service is often provided in conjunction with parking and loaning.
Title transfer—A service in which changes in ownership of a specific gas package are recorded
by the market center. Title may transfer several times for some gas before it leaves the center.
Electronic trading —Trading systems that either electronically match buyers with sellers or
facilitate direct negotiation for legally binding transactions.
Administration—Assistance to shippers with the administrative aspects of gas transfers, such as
nominations and confirmations.
Compression—Provision of compression as a separate service. If compression is bundled with
transportation, it is not a separate service.
Risk Management—Services that relate to reducing the risk of price changes to gas buyers and
sellers, for example, exchanges of futures for physicals.
Hub-to-hub transfers—Simultaneous receipt of a customer’s gas into a connection associated with
one center and an instantaneous delivery at a distant connection associated with another center.
Source: Energy Information Administration (2003).
© 2006 by Taylor and Francis Group, LLC
256 Fundamentals of Natural Gas Processing
operate near ambient pressure and (2) much larger underground facilities
(depleted oil and gas fields, salt caverns, and aquifers) that operate at elevated
pressures. Overall, underground storage is more important and the discussion
here is limited to this topic. The use of liquefied natural gas (LNG) storage for
The United States Geological Survey first proposed use of underground
storage for natural gas in 1909. The first North American facility was a depleted
gas reservoir in Welland County, Ontario, Canada converted to storage use in
1915. This facility was followed by the Zoar field in Concord, New York in 1916
(Cates, 2001). In 1916, a German patent was issued on solution mining of salt
caverns for storage of crude oil and distillates, and in 1950, this concept was
applied to storage of NGL in the Keystone field, West Texas. The first use of a
salt cavern for natural gas storage came in 1961, in Marysville, Michigan, and
the first use of a leached salt dome for storage was in 1964, at Eminence,
Mississippi (Beckman and Determeyer, 1997). The number of storage facilities
of underground facilities in the United States in 2002. In the 2000 to 2003 period,
the total underground storage capacity has remained constant at around 8.2 Tcf
(220 Bm
3
) (Energy Information Administration, 2005c).
Underground storage is prevalent throughout the world, with two exceptions,
Japan and Korea. In these two countries, gas storage is primarily in the form of
LNG. (International Energy Agency, 2002).
Aquifers are underground natural water reservoirs that can, under the right
circumstances, be used for gas storage. However, aquifer storage is usually the
most expensive and, thus, the least desirable underground storage method for six
reasons (NaturalGas.org, 2004):
1. Geologic characteristics of a specific aquifer are generally not well
known, which is usually not the case with a depleted gas or oil field,
FIGURE 12.3 United States natural gas demand over 3-year period (Energy Information
Administration, 2005b).
1.0
Jan-02 Jan-03 Jan-04 Jan-05 Jul-02 Jul-03 Jul-04
1.2
1.4
1.6
1.8
2.0
2.2
2.4
2.6
2.8
3.0
M
o
n
t
h
l
y

g
a
s

d
e
m
a
n
d
,

T
c
f

has grown steadily in recent decades; Figure 12.4 shows the capacity and number
peak shaving is discussed in Chapter 13.
© 2006 by Taylor and Francis Group, LLC
Transportation and Storage 257
and, consequently, considerable resources must be expended to deter-
mine the suitability of the aquifer for gas storage.
2. Infrastructure (wells, pipelines, dehydration facilities, compression
equipment, etc.) is unavailable at the aquifer site, whereas a depleted
gas reservoir would have most of this infrastructure in place.
3. Considerable injection pressure may be required to displace the water
with gas.
4. Withdrawn gas requires dehydration.
5. Aquifer formations generally require a much higher level of cushion
or base gas (up to 80% of the total gas volume) than do depleted fields
or salt caverns, and, thus, less of the reservoir volume is usable.
6. Environmental regulations govern the use of aquifers for gas storage.
All of the above factors increase both the capital cost and time necessary for
development of aquifer storage. Consequently, depleted fields and salt caverns
are normally preferred.
A common and relatively inexpensive technique for creating large storage facilities
is solution mining of underground salt beds. After the salt bed has been located and
the appropriate well or wells drilled, a coaxial pipe is inserted in the well bore. Water
is then pumped down the annulus of the pipe, and the dissolved brine is withdrawn
that would permit gas leaks and is well suited for pressurized gas storage. Beckman
and Determeyer (1997) present additional information on this subject.
In common with all engineering processes, storage gas reservoirs have their
FIGURE 12.4 Capacity of underground storage facilities in the United States at the end
of 2004 (Energy Information Administration, 2005c).
Salt caverns
30 sites
30 Bcf
Aquifers
43 sites
1,237 Bcf
Depleted fields
318 sites
6,734 Bcf
own terminology. Table 12.2 summarizes the most important terms.
through the inner pipe (Figure 12.5). The cavern formed tends to be free from fractures
© 2006 by Taylor and Francis Group, LLC
258 Fundamentals of Natural Gas Processing
FIGURE 12.5 Generating underground-storage salt caverns by leaching with fresh water.
(Adapted from Beckman and Determeyer, 1997.)
TABLE 12.2
Important Terminology for Underground Storage
Total gas storage capacity Maximum volume of gas that can be stored in an underground
storage facility
Total gas in storage Volume of storage in the reservoir at a particular time
Cushion gas (or base gas) Volume of gas intended as a permanent inventory in a storage
reservoir to maintain adequate pressure and deliverability rates
throughout the withdrawal season
Working gas capacity Total gas storage capacity minus cushion gas
Deliverability Most often expressed as measure of the amount of gas that can be
delivered (withdrawn) from a storage facility on a daily basis;
deliverability is variable and depends on such factors as the amount
of gas in the reservoir at any time, the reservoir pressure, and
compression capability available to the reservoir
Injection capacity (or rate) Amount of gas that can be injected into a storage facility on a daily
basis; injection rate is also variable; depends on the same factors
as deliverability.
Source: Energy Information Administration (2004).
Brine
Fresh
water
Cap rock
Salt dome
Leached
cavity
© 2006 by Taylor and Francis Group, LLC
Transportation and Storage 259
The measures above are not absolute and are subject to change and interpre-
tation. For example, in practice, a storage facility may be able to exceed certified
total gas storage capacity by exceeding certain operational parameters. Also, the
distinction between cushion and working gas is somewhat arbitrary, so facilities
can withdraw some cushion gas for supply to markets in times of heavy demand.
Figure 12.6 shows the total gas in storage and the working gas for the 2002
over the same time period. Note that even during the winter peak demand periods,
some gas is being injected.
More detailed information on gas storage in depleted reservoirs and aquifers
is available in Katz and Lee (1990).
12.3 LIQUIDS
12.3.1 TRANSPORTATION
In common with other petroleum products, liquids are transported by dedicated
pipeline, by batching with other liquids in a petroleum pipeline, and by rail, truck,
and marine transport. The most feasible and economical means of transport
depends upon many factors, including:
• Product specifications, especially with respect to water and sulfur
requirements
• Existing infrastructure
• Volatility
• Rate of product produced
FIGURE 12.6 Underground gas storage volumes in the United States, 2002 through 2004.
The lower solid line denotes the cushion gas volume and upper dashed line represents the
total storage capacity (Energy Information Administration, undated).
4
4.5
5
5.5
6
6.5
7
7.5
8
8.5
N
a
t
u
r
a
l

g
a
s

s
t
o
r
a
g
e

T
c
f

Jan-02 Jan-03 Jan-04 Jan-05 Jul-02 Jul-03 Jul-04
to 2004 period. The pattern follows that for the demand shown in Figure 12.3.
Figure 12.7 shows the total gas injected and withdrawn and the net gas withdrawn
© 2006 by Taylor and Francis Group, LLC
260 Fundamentals of Natural Gas Processing
• Distance between plant and customer
• Geographic location
The first factor dictates the limitations of pipeline or storage container usage.
Bulk containers and pipelines have different contamination problems. For con-
tainers, the possibility exists of a residual “heel” when one fluid is replaced with
another. Even with a minimal heel, surface adsorption of water and sulfur com-
pounds can contaminate the product being loaded. Many tank car loads are
contaminated in these two ways.
In pipelining a fluid, contamination between different products transported
in the same line is impossible to avoid. An interfacial zone exists between the
two fluids, where mixing occurs because of convection. Pipeline operators offload
the mixture, called transmix, into storage tanks at pipeline terminals. The transmix
then is fractionated to regain the two products. To reduce the amount of transmix
by up to 50%, pipeline operators use batching pigs (Webb, 1978).
Whereas pipeline operators know how much transmix to remove from the
shipment, they have no idea of how much product can be contaminated by surface
adsorption. Even worse, some fluids are shipped saturated with water. If the fluid
temperature drops, water may condense and collect in line depressions. Large
quantities of water-dry product can be contaminated this way. Therefore, pipeline
operators try to monitor product quality when contamination can occur. When it
happens, the pipeline goes out of service and “chasing fluids” are run through
the line to remove contaminates. This process is costly to both the pipeline
operator and those who use the pipeline.
FIGURE 12.7 Volume of gas injected () and withdrawn (᭛) and net gas withdrawn ()
from underground storage on monthly basis between January 2002 and December 2004
(Energy Information Administration, undated).
−0.6
−0.4
−0.2
0.0
0.2
0.4
0.6
0.8
1.0
M
o
n
t
h
l
y

g
a
s

v
o
l
u
m
e
s
,

T
c
f
Jan-02 Jan-03 Jan-04 Jan-05 Jul-02 Jul-03 Jul-04
© 2006 by Taylor and Francis Group, LLC
Transportation and Storage 261
Because of the diverse nature of the liquid products from a gas plant, we
address transporting each of the products separately.
12.3.1.1 Natural Gas Liquids
Natural gas liquids (NGL) are the major liquid product from gas plants that do
not further fractionate demethanizer bottoms. Usually, the quantity is large enough
to dedicate a pipeline to the customers, who typically are operators of central
NGL fractionation and storage hubs. Specifications exist on methane content,
water content, and sulfur content, but the need for continuous flow is the major
reason for the dedicated line.
12.3.1.2 Ethane-Propane Mixtures
The high vapor pressure of ethane−propane mixtures (E-P mix) and water-content
specification make dedicated pipelines the common transport means for this
liquid. Common carriers in Texas provide dedicated pipelines to transport E-P
mixes to the major customers, refineries, and chemical plants. This also applies
to high purity ethane.
12.3.1.3 Liquefied Petroleum Gas
Because it is a widely used fuel with strict specifications on water and sulfur,
liquefied petroleum gas, or LPG, (more specifically HD-5 propane) must use
dedicated transport. During the early years of the evolving LPG transportation
system, the principal method was the railroad tank car, with pipelines playing a
minor role, but by 1962, more than 40% of LPG was transported by pipeline and
today it is the preferred method. A major reason for this shift was the construction
of the Mid-America Pipeline in the early 1960s, a line added to serve the upper
Midwest. This 2,200 mile (3,500 km) pipeline starts in eastern New Mexico and
runs up to Wisconsin and Minnesota. Since then, additional LPG pipelines have
been added to service most of the eastern United States.
12.3.1.4 Butanes
No strong preference exists for how butanes are shipped from gas plants, because
the quantities produced usually do not justify the cost of a pipeline. If available,
railroad tank car is desirable. Butane is usually stored on-site to permit batch
loading into tank cars or trucks to a hub. For distribution of the product long
distances from the hub, it can be put into pipelines that transport gasoline and
diesel. Contamination is not an issue in this case, because the butanes are normally
blended with gasoline to adjust the gasoline vapor pressure.
12.3.1.5 Natural Gasoline
Natural gasoline production is typically so small that it is stored and then loaded
into a tank car or truck. There is little contamination issue because the customer
is usually a refinery which will use it for gasoline blending.
© 2006 by Taylor and Francis Group, LLC
262 Fundamentals of Natural Gas Processing
12.3.2 STORAGE
Light gas-plant producer liquids can be stored aboveground in pressurized vessels
or below ground in large caverns. Figure 12.8 shows a typical row of cylindrical
storage vessels, and Figure 12.9 is an example of a spherical vessel for butane
storage (Cannon, 1993). Additional information on cavern storage is available in
the book by Katz and Lee (1990).
FIGURE 12.8 Above ground storage vessels for LPG and NGL.
FIGURE 12.9 Spherical storage tanks. (Courtesy of Duke Energy Field Services.)
storage. Figure 12.10 shows an underground cavern being prepared for NGL
© 2006 by Taylor and Francis Group, LLC
Transportation and Storage 263
REFERENCES
Beckman, K.L. and Determeyer, P.L, Natural gas storage: historical development and
expected evolution, GasTIPS, 3, 13, 1997.
Cates, H.C., Storage functions evolve to match changes in U.S. natural gas industry, Oil
Gas J., 99.41, 48, 2001
Cannon, R.E., The Gas Processing Industry, Origins and Evolution, Gas Processors Asso-
ciation, Tulsa, OK, 1993, 245.
Energy Information Administration, U.S. Department of Energy, Natural Gas Navigator,
FIGURE 12.10 Underground storage cavern being prepared for NGL service (Cannon,
1993). (Courtesy of Gas Processors Association.)
U.S. Natural Gas Gross Withdrawals and Production, undated http://tonto.eia.
doe.gov/dnav/ng/ng_prod_sum_dcu_NUS_a.htm, Retrieved September 2005.
© 2006 by Taylor and Francis Group, LLC
264 Fundamentals of Natural Gas Processing
Energy Information Administration, U.S. Department of Energy, Office of Oil and Gas, Natural
2005.
Energy Information Administration, U.S. Department of Energy, The Basics of Underground
Energy Information Administration, U.S. Department of Energy, Office of Oil and Gas,
Changes in U.S. Natural Gas Transportation Infrastructure in 2004, 2005a,
Retrieved September 2005.
Energy Information Administration, U.S. Department of Energy, Natural Gas Consumption by
Energy Information Administration, U.S. Department of Energy, Underground Natural Gas
Gray, L.A., Hong, W., and Naxin, M. Construction of major Chinese gas pipeline enters
final phase in 2003, Oil Gas J., 101.7, 68, 2003.
International Energy Agency, Natural Gas Information 2002, Paris, 2002.
Katz, D.L. and Lee, R.L., Natural Gas Engineering, Production and Storage, McGraw-
Hill, New York, 1990.
Manning, F.S. and Thompson, R.E., Oilfield Processing of Petroleum, Pennwell Publishing,
Tulsa, Oklahoma. 1991.
2005.
Smith, C.E., Surging U.S., steady Asia-Pacific lead construction plans, Oil Gas J., 104.6,
57, 2006.
Smith,C.E., True, W.R., and Stell, J., U.S. Gas carriers see 2004 net jump: construction
plans rebound, Oil Gas J. 103.34, 50, 2005.
True, W.R., Construction plans surge on prospects for gas use, Oil Gas J., 102.3, 58, 2004.
True, W.R. and Stell, J., U.S. construction plans slide: pipeline companies experience flat
2003, continue mergers, Oil Gas J., 102.32, 52, 2004.
Webb, B.C., Guidelines set out for pipeline plugging, Oil Gas J., 76, 196, 1978.
Natural Gas Storage, 2004, http://www.eia.doe.gov/pub/oil_gas/natural_gas/analysis
_publications/storagebasics/storagebasics.html, Retrieved October 2005.
Gas Market Centers and Hubs: a 2003 update, 2003, www.eia.doe.gov/pub/oil_gas/
natural_gas/feature_articles/2003/market_hubs/mkthubs03.pdf, Retrieved September
www.eia.doe.gov/pub/oil_gas/natural_gas/feature_articles/2005/ngtrans/ngtrans.pdf,
NaturalGas.org, Storage of Natural Gas, 2004, WWW.naturalgas.org, Retrieved September
Retrieved September 2005.
End Use, 2005b, http://tonto.eia.doe.gov/dnav/ng/xls/ng_cons_sum_dcu_nus_m.xls,
Storage Capacity 2005c, http://tonto.eia.doe.gov/dnav/ng/xls/ng_ stor_ cap_ dcu_
nus_a.xls, Retrieved September 2005.
© 2006 by Taylor and Francis Group, LLC
265
13
Liquefied Natural Gas
13.1 INTRODUCTION
The rationale for liquefying natural gas is simple: at atmospheric pressure,
the liquid density at the normal boiling point of methane is approximately
610 times greater than that of the gas at ambient temperature and pressure.
Consequently, a given volume of liquid contains over 600 times the heating
Field Liquids Removal
Gas & Liquids from Wells
Inlet
Compression
Sulfur
Recovery
Gas Treating
Dehydration
Nitrogen
Rejection
Helium
Recovery
Outlet
Compression
Elemental
Sulfur
Field Dehydration
Field Acid Gas Removal
Field Compression
Sales Gas
Water &
Solids
Natural
Gas Liquids
Natural
Gasoline
Raw Helium
CO
2
Liquefied
Natural Gas
Inlet Receiving
Hydrocarbon
Recovery
Liquids Processing
Liquefaction
© 2006 by Taylor and Francis Group, LLC
266 Fundamentals of Natural Gas Processing
value as the same volume of ambient gas. This density increase at ambient
pressure makes it attractive to liquefy, transport, and store natural gas in large
quantities and makes technically feasible the transport of the equivalent of
several Bcf of gas per ship load from “stranded” gas fields to markets.
Liquefaction and transport becomes economically feasible when the size of
the reserves justify the capital investment of a liquefied natural gas (LNG)
plant. Storage applications include storage at LNG terminals and, just as
important, storage for peak shaving operations of gas utilities.
After a general discussion of peak shaving and baseload plants, this chapter
focuses on eight topics:
1. Gas treating before liquefaction
2. Liquefaction cycles
3. Storage
4. Transport
5. Regasification and cold utilization
6. Economics
7. Plant efficiency
8. Safety and environmental considerations
13.1.1 PEAK SHAVING PLANTS AND SATELLITE FACILITIES
need for gas storage facilities. Because natural gas fields are generally located
far from residential and industrial consumers, storing large quantities of gas
near the point of consumption to supplement the normal supply of pipeline
discussed gas storage techniques, the most common of which is underground
storage in depleted oil or gas fields, salt caverns, or abandoned mines that can
be effectively sealed. Where underground storage is unavailable, aboveground
storage of natural gas as LNG becomes attractive, and utilities use relatively
small liquefaction, storage, and regasification plants to meet the demand. Plants
that combine all three of these tasks are referred to as peak shaving plants.
“Stranded” utilities, those not connected to the national pipeline grid, rely
upon LNG received by truck to support their customers. Any LNG facilities
like these, which contain only storage and regasification units, are called
“satellite facilities.”
shaving facility. Gas treating and compression are discussed in earlier chapters.
This chapter discusses liquefaction, liquid storage, and regasification. Odorant
injection may or may not be required at the peak shaving plant.
The first peak shaving plant built in the United States was in Cleveland, Ohio, in
1941 (Miller and Clark, 1941). Although the plant performed successfully for several
years, in October 1943, a metallurgical failure in a storage tank resulted in a
In the previous chapter on Transportation and Storage, Figures 12.3, 12.6, and
Figure 13.1 shows a block diagram of the common steps involved in a peak
gas during periods of peak demand (peak shaving) is essential. Chapter 12
12.7 clearly show the large seasonal shifts in gas demand that result in the
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 267
fire and explosion (GAO, 1978) that destroyed the plant, with a heavy loss of life.*
Although this disaster was a major setback to the industry, in 2004 the United States
had 59 peak shaving plants, 39 satellite facilities, four LNG marine-import terminals,
and one LNG marine-export terminal (Energy Information Administration, 2004).
13.1.2 BASELOAD PLANTS AND STRANDED RESERVES
Baseload plants exist to provide the industrial world with gas from stranded reserves
in remote places. Stranded gas reserves are located where no economic use for the
natural gas exists at the point of origin and where transportation of the gas by pipeline
to a point of end use is not feasible. Romanow (2001) estimates that approximately
60% of the world’s gas reserves are considered stranded. When compressed gas
pipelines are impractical or impossible, a limited number of conventional options are
open (Taylor et al., 2001), such as compression and transport of the gas in specially
built ships (Wagner, 2002), conversion of the natural gas into a liquid through gas-
to-liquid (GTL) technology, and liquefaction and shipment of the gas in specially built
LNG vessels. Leibon et al. (1986) as well as Taylor et al. (2001) evaluate the status
of several of the technologies. Hidayati et al. (1998) compare the cost of a compressed
gas pipeline to LNG carriers for a large Indonesian project. Some unconventional
methods that have been considered include conversion of the natural gas to hydrates
for shipping (Gudmundsson and Mork, 2001) and even use of a train of airships that
contain natural gas. Presently, LNG is the most viable option in almost all situations
FIGURE 13.1 Schematic of peak-shaving facility.
* The Energy Information Agency (2004) states that “Modern LNG plants are designed and constructed
in accordance with strict codes and standards that would not have been met by the Cleveland plant.
LNG safety standards have been issued by the National Fire Protection Agency in NFPA 59A.”
Regasification
Odorant
injection
Gas treating
Storage
N
a
t
u
r
a
l

g
a
s

p
i
p
e
l
i
n
e
H
2
S, CO
2
and odorant
H
2
O
Compression
liquefaction
Heavy
hydrocarbons
© 2006 by Taylor and Francis Group, LLC
268 Fundamentals of Natural Gas Processing
involving stranded reserves, if the gas can be pipelined to a seaport. However, to
economically justify a traditional baseload LNG plant requires reserves of approxi-
mately 3 Tcf (80 Bm
3
). Newer designs have reduced the reserve volumes down to
around 1 Tcf (30 Bm
3
) (Price et al., 2000).
As Figure 13.2 shows, bringing the gas from the field to the customer involves
four steps (Energy Information Administration, 2003b):
1. Gas production, gathering, and processing
2. LNG production, including gas treating, liquefaction, NGL condensate
removal, and LNG storage and loading
3. LNG shipping
4. LNG receiving facilities, which include unloading, storage, regasifica-
tion, and distribution
Depending on the specific situation, not all plants will have all the processes shown,
and some plants may have additional processes.
countries, Indonesia, Malaysia, Qatar, and Algeria, accounted for 64% of the
world’s exports, and two countries, Japan and South Korea, accounted for 63%
of the world’s imports (Energy Information Administration, 2003b).
FIGURE 13.2 Schematic of a baseload plant combined with transporting, receiving, and
regasification facilities.
Field
operations
Ship Loading
Nitrogen
rejection
H
2
S/CO
2
NGL
Compression
Gas treating
Hydrocarbon recovery
N
2
Ship
Offloading
Storage
Regasification Distribution
Compression
Liquefaction
Storage
H
2
O
Figures 13.3 and 13.4 show the worldwide movements of LNG. In 2002, four
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 269
States compared with the Henry Hub* price. The LNG price is higher or equal
to that of Henry Hub, with the exception of the 2000 price. This parity results in
part from the fact that most of the LNG importation facilities are on the East
Coast of the United States and do not have to incur pipeline transportation fees
from Henry Hub to highly populated end markets.
The liquefaction and storage facilities for baseload plants are necessarily quite
price through liquefaction, marine transportation, and regasification) for three
existing LNG operations. In all three situations, the price of the stranded gas is
FIGURE 13.3 Worldwide LNG exports in 2002. Total exports were equivalent to 5.37
Tcf (152 Bm
3
) of natural gas (Energy Information Administration, 2003b).
centers.
FIGURE 13.4 Worldwide LNG imports in 2002 (Energy Information Administration, 2003b).
Indonesia
20.6%
Algeria
17.4%
Qatar
12.6%
Malaysia
13.8%
Remaining
countries
35.6%
Japan
48%
France
10%
South Korea
15%
Remaining
countries
27%
large, complex, and expensive. Table 13.1 lists the plants in operation in 2003.
Table 13.2 lists an approximate breakdown of costs (stranded wellhead gas
Figure 13.5 shows the trend of the price for LNG imported into the United
* Henry Hub is one of a number of large marketing centers. See Chapter 12 for a discussion of these
© 2006 by Taylor and Francis Group, LLC
270 Fundamentals of Natural Gas Processing
less than 40% of the total cost of marketing the reserve. Cost distribution that
includes offshore facilities, if they are developed, would be different.
The worldwide future of the LNG industry utilization of stranded reserves
countries in different regions of the world accounted for an increase of over 60%
in the capacity available in 1999.
Stranded gas exists offshore as well and can be either associated gas, with
oil production, or nonassociated gas. Two papers, Finn, et al. (2000) and Wagner
and Cone (2004), look at possible LNG liquefaction cycles for offshore applica-
tions. Finn et al. (2000) compare the various liquefaction cycles that might apply
offshore. Wagner and Cone (2004) conclude that if the gas is nonassociated,
reserve sizes comparable to those for onshore baseload plants will be needed.
Associated gas might be more economical at lower reserve volumes because the
value of the combined gas−oil project must be considered. They note the interest
in having offshore liquid storage facilities alongside the LNG plant.
13.2 GAS TREATING BEFORE LIQUEFACTION
Production of LNG requires temperatures as low as −258°F (−161°C), the normal
boiling point of methane, and, consequently, the allowable impurity levels in a
gas to be liquefied are much lower than that of a pipeline-quality gas. For example,
gas to be pipelined and sold to residential or industrial customers may contain a
maximum of 3 to 4 mol% carbon dioxide, but gas for liquefaction should have
specifications for the two cases.
Obviously, gas processed for LNG must have much more aggressive removal
of water, nitrogen, and carbon dioxide than does gas destined for pipelines. The
tight specifications on all the above components, except for nitrogen and mercury,
are needed to avoid solids deposition that will plug the heat exchangers. Nitrogen
FIGURE 13.5 Price of imported LNG (

) compared with Henry Hub price (- - ) (Energy
Information Administration, 2003b).
1.5
2.5
3.5
4.5
5.5
1992 1994 1996 1998 2000 2002
Year
C
o
n
s
t
a
n
t

2
0
0
2

U
.
S
.

$
/
M
M
B
t
u
appears bright. Table 13.3 lists the projects that are under construction. Seven
a carbon dioxide content of less than 50 ppmv. Table 13.4 compares compositional
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 271
is a volatile diluent which, at higher concentrations, can raise the potential for
stratification and rollover (discussed in section 13.4). Elemental mercury presents
of mercury condense in the cryogenic heat exchangers and form an amalgam with
aluminum that can lead to exchanger failure. Consequently, mercury must be
removed to a level of 0.01 mg/Nm
3
. Techniques for achieving the desired levels for
all of the above mentioned impurities are discussed in the previous chapters.
TABLE 13.1
Baseload LNG Plants
Plant Country
Capacity
Bcf gas /yr
(BSm3/yr)
Number of
Trains
Start-up
Date
Arzew GL1Z Algeria 385 (10.9) 6 1978
Arzew GL2Z Algeria 404 (11.4) 6 1981
Arzew (Camel)
GL4Z
Algeria 44 (1.2) 3 1964
SkikdaGL1K phase
I and II
a
Algeria 292 (8.27) 6 1972/1981
Marsa el Brega Libya 29 0.82) 3 1970
Bonny Island Nigeria 321 (9.09) 2 1999
Bonny Island
Train 3
Nigeria 141 (3.99) 1 2002
Withnell Bay Australia 365 (10.3) 3 1989
Lumut 1 Brunei 351 (9.93) 5 1972
Arun Phase
I, II, and III
Indonesia 331 (9.37) 4 1978/1984/1986
Bontang A-H Indonesia 1,101 (31.2) 8 1977
Bintulu MLNG 1 Malaysia 370 (10.5) 3 1983
Bintulu
MLNG 2
Malaysia 380 (10.7) 3 1994
Bintulu
MLNG 3
Malaysia 166 (4.70) 1 2003
Qalhat Oman 356 (10.1) 2 2000
Ras Laffan Qatar 404 (11.4) 3 1996
Ras Laffan Qatar 321 (9.09) 2 1998
Das Island I, II UAE 278 (7.87) 3 1977/1994
Point Fortin Trinadad
and Tobago
482 (13.6) 3 1999/2003
Kenai United States 68 (1.92) 2 1969
World total 6,589 (186.6) 66
a
An explosion and fire occurred January, 2004 that killed 27, injured 70, and shut down a
Source: Energy Information Administration (2003b).
serious problems in cryogenic operations. As noted in Chapter 9 trace quantities
portion of the plant (for details, see California Energy Commission [2004]).
© 2006 by Taylor and Francis Group, LLC
272 Fundamentals of Natural Gas Processing
These compositions are for the LNG produced and not the feed gas to the plant.
13.3 LIQUEFACTION CYCLES
The two most common methods that have been used in engineering practice to
produce low temperatures are Joule-Thomson expansion and expansion in an
engine doing external work. This section discusses each of these processes in
detail and analyzes them thermodynamically.
13.3.1 JOULE-THOMSON CYCLES
The Joule-Thomson coefficient is the change in temperature that results when a gas
is expanded adiabatically from one constant pressure to another in such a way that
no external work is done and no net conversion of internal energy to kinetic energy
of mass motion occurs. Thermodynamically, it is an irreversible process that wastes
the potential for doing useful work with the pressure drop. However, it is as simple
The thermodynamic definition of the Joule-Thomson coefficient is
. (13.1)
One of the more important thermodynamic relations that involves the Joule-
Thomson coefficient is
. (13.2)
TABLE 13.2
Cost Breakdown in $/MM Btu for LNG Plants
Project Wellhead
Gas
Liquefaction Transportation Regasification Minimum
CIF Cost
a
Qatar 0.50−0.75 0.40−0.60 1.10−1.20 0.40−0.60 2.45
North West Shelf
(Australia)
0.65−0.95 0.40−0.60 0.75−0.95 0.35−0.55 2.15
Bontang,
Indonesia
0.60−0.80 0.45−0.65 0.55−0.75 0.30−0.60 1.90
a
CIF Carriage + insurance + freight (i.e., cost delivered to customer).
Source: Troner (2001).
µ

( )
T
P
h
µ




− −
( )
|
.

`
}


( )
1 1
C
V T
V
T
C
H
P
P
P
P
T
Table 13.5 shows the range of compositions and properties for 17 LNG samples.
as a valve or orifice and finds wide use in refrigeration cycles (see Chapter 7).
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 273
Combination of the above relation with the ideal gas law (PV RT) gives
m 0, and thus no temperature change occurs when an ideal gas undergoes a
Joule-Thomson expansion. For a real gas, the Joule-Thomson coefficient may be
positive (the gas cools upon expansion), negative (the gas warms upon expansion),
or zero. The locus of all points on a pressure−temperature plot where the Joule-
that the Joule-Thomson inversion curve for methane expansions must take place
below the curve to produce refrigeration.
The behavior of several gases upon expansion from 101 bar (1,470 psia) to 1 bar
and nitrogen, the cooling effect upon expansion when started at ambient temperature
TABLE 13.3
LNG Liquefaction Facilities in Planning, Engineering,
and Construction Phases
Country Location
Capacity
Million tpy
a
Comments
Angola Soyo 5.0 Planning stage
Algeria Arzew 5.2 Planning stage
Australia Barrow Island 5.0 Planning stage
Australia Darwin 3.0 Under construction
Australia Karratha 3.2 Engineering
Egypt Idku 3.6 Under construction
Equatorial Guinea Bioko Island 3.4 Under construction
Indonesia Berau Bay, Papua 7.0 Engineering
Nigeria Brass Terminal 10.0 Engineering
Nigeria Bonny Island 4.0 Under construction
Nigeria Bonny Island 3.0 Engineering
Nigeria Bonny Island 8.8 (2 projects) Planning stage
Norway Melkoya Island 4.0 Under construction
Oman Qalhat, sur 3.3 Under construction
Peru Pampa Melchorita 4.0 Planning stage
Qatar Ras Laffan 12.5 (2 projects) Under construction
Qatar Ras Laffin 7.8 Engineering
Qatar Ras Laffin 7.8 Planning stage
Russia Prigorodnoye Sakhalin 9.6 Under construction
Trinidad and Tobago Point Fortin 4.6 Under construction
Trinidad and Tobago Point Fortin 6.0 (2 projects) Planning stage
Venezuela Gran Marischal
De Ayacucho Sucre
6.8 (2 projects) Planning stage
Yeman Bal Haf 6.2 Planning stage
a
tpy denotes metric tons per year.
Source: Adapted from LNG Observer (2005).
(14.5 psia) is shown in Table 13.6. Two items should be noted. First, for both methane
Thomson coefficient is zero is known as the inversion curve. Figure 13.6 shows
© 2006 by Taylor and Francis Group, LLC
274 Fundamentals of Natural Gas Processing
(80°

F, 27°C) is relatively small. Second, the cooling effect increases significantly as
the initial temperature is lowered. For helium, the expansion results in heating the
gas rather than cooling. The temperature increase remains constant because the Joule-
Thomson coefficient remains nearly constant over the temperature range considered.
TABLE 13.4
Compositional Specifications on Feed to LNG Plant and on Pipeline Gas
Impurity Feed to LNG Plant
a
Pipeline Gas
b
Water < 0.1 ppmv
c
150 ppmv, (7.0 lb/MMscf,
110 kg/Sm
3
)
Hydrogen sulfide < 4 ppmv 0.25 − 0.30 gr/100 scf
(5.7 − 22.9 mg/Sm
3
)
Carbon dioxide < 50 ppmv 3 to 4 mole%
Total sulfur
(H
2
S, COS, organic sulfur)
< 20 ppmv 5 − 20 gr/100 scf
(115 − 459 mg/Sm
3
)
Nitrogen <1 mol% 3 mol%
Mercury < 0.01 mg/Nm3
Butanes 2 mol% max
Pentanes+ 0.1 mol% max
Aromatics < 2 ppmv
c
a
Foglietta (2002).
b
Engineering Data Book (2005a).
c
McCartney (2005).
TABLE 13.5
Typical LNG Compositions
Component Composition Range (mol%)
Nitrogen 0.00 – 1.00
Methane 84.55 – 96.38
Ethane 2.00 – 11.41
Propane 0.35 – 3.21
Isobutane 0.00 – 0.70
n-Butane 0.00 – 1.30
Isopentane 0.00 – 0.02
n-Pentane 0.00 – 0.04
HHV gas
Btu/scf (kJ/Sm
3
)
1021 – 1157
(38,000 – 43,090)
Wobbe number 1353 – 1432
GPM, on C
2
+ basis
(m
3
/1,000m
3
)
0.71 – 4.08
(0.094 – 0.543)
Source: McCartney (2003).
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 275
Because methane, the principal constituent of natural gas, must be cooled to
−258°F (−161°C) before it becomes a liquid at 1 atmosphere pressure, a liquefier
that uses only a Joule-Thomson expansion requires more than a compressor and
an expansion valve if it is to function at reasonable initial pressures. A counterflow
heat exchanger needs to be added to make a complete system. A simple Joule-
The liquefaction cycle begins with natural gas being compressed and sent
through the heat exchanger and expansion valve. Upon expansion, the gas cools
(approximately 84°F [47°C] if the gas is principally methane and the expansion
is from 1,500 to 14.7 psia [101 to 1 bar]), but none liquefies because a temperature
FIGURE 13.6 Joule-Thomson inversion curve for pure methane as a function of temper-
ature and pressure (Goodwin, 1974).
TABLE 13.6
Expansion from 1470 psia (101 bar) to 14.5 psia (1 bar)
Initial Temperature Final Temperature t
final
−t
initial

؇F (؇C) ؇F (؇C) ؇F (؇C)
Methane 80 ( 27) −4 (−20) −44 (−47)
Nitrogen 80 ( 27) 46 ( 8) −34 (−19)
Helium 80 ( 27) 91 ( 33) 11 ( 6)
Methane −10 (−23) −125 (−87) −115 (−64)
Nitrogen −10 (−23) −60 (−51) −50 (−28)
Helium −10 (−23) 1 (−17) 11 ( 6)
Methane −46 (−43) −215 (−137) −169 (−94)
Nitrogen −46 (−43) −107 (−77) −61 (−34)
Helium −46 (−43) −35 (−37) 11 ( 6)
0
2000
4000
6000
8000
10000
−200 0 200 400 600
Temperature, °F
P
r
e
s
s
u
r
e
,

p
s
i
a
Thomson system suitable for natural gas liquefaction is shown in Figure 13.7.
© 2006 by Taylor and Francis Group, LLC
276 Fundamentals of Natural Gas Processing
drop of approximately 338°F (188°C) is required to convert the gas to a liquid.
Thus, all of the chilled low-pressure gas is recycled through the heat exchanger
for recompression. This cold low-pressure gas lowers the temperature of the high-
pressure gas stream ahead of the expansion valve, which results in a lower
temperature upon expansion.
As long as all of the gas being expanded is recycled through the counterflow
heat exchanger to cool the high-pressure gas stream, temperatures will be pro-
gressively lower upon expansion. The process continues until liquid is formed
during the expansion from high to low pressure. The liquid formed is separated
from the low-pressure gas stream in the liquid receiver and is ultimately with-
drawn as the product. The amount of low-pressure gas recycled to the compressor
is now significantly reduced, which cuts back on the cooling effect in the heat
exchanger. With the addition of makeup gas to the low-pressure side of the
compressor to compensate for the liquid product being withdrawn, a steady-state
is reached in the liquefaction system and no further cooling can be achieved.
FIGURE 13.7 Simple Joule-Thomson liquefaction cycle.
ermodynamic
boundary
Heat
exchanger
J-T
valve
Compressor
Air-cooled
heat exchanger
LNG product
Methane make-up
gas
q
L
1
3
4
2
5
6
Liquid
receiver
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 277
The first law of thermodynamics* for a steady-state flow system is
, (13.3)
where PE and KE are, respectively, the potential and kinetic energy per unit
mass. The enthalpy, h, heat term, q, and work term, w
S
, are on a mass basis, and
represents the mass flow rate. Application of the equation to the components
valve, and liquid receiver) gives the relation
, (13. 4)
where the overall enthalpy change of the gas, h, on a mass basis equals the
heat leak, , per unit mass of gas. On a per unit of mass flow of entering gas,
and defining , the fraction of entering gas withdrawn as a liquid, the
equation becomes:
(13.5)
or
(13.6)
For a given system, h
2
, h
3
, and q
L
are essentially fixed, so the only way to increase
liquefaction is to decrease the inlet gas enthalpy, h
1
, which is done by increasing the
inlet pressure, assuming that the compressor outlet gas temperature remains constant.
Thus, more compressor work should lead to more liquid production.
Example 13.1 Methane is to be liquefied in a Joule-Thomson cycle as shown in
Figure 13.7. The methane enters the heat exchanger at 80°F and 1,500 psia and
expands to 14.7 psia.
1. Calculate the fraction of methane entering the system that is liquefied.
2. Estimate the % decrease in production if a heat leak of 15 Btu/lb of methane
entering is present and if a temperature approach of 5°C is obtained at the warm
end of the exchanger.
3. Calculate the fraction liquefied if the pressure is 2,000 psia.
* The first law equation for open, steady-flow systems is developed in most chemical and mechanical
engineering textbooks. For example, see Smith et al. (2001).
0 − + + + − ∆[( ) ] h KE PE m mq mw
S

m
∆h q
L

q
L
f m m
1 2
/
f h f h h q
L 2 3 1
1 + − − ( )
f
h h q
h h
L

− −

3 1
3 2
q
L
inside the thermodynamic boundary of Figure 13.7 (heat exchanger, Joule-Thomson
© 2006 by Taylor and Francis Group, LLC
278 Fundamentals of Natural Gas Processing
Calculate the fraction liquefied—An ideal heat exchanger is assumed (no warm
end and no pressure drop). From the methane pressure−enthalpy diagram and
h
1
350 Btu/lb (80°F, 1500 psia)
h
3
392 Btu/lb (80°F, 14.7 psia)
h
2
0 Btu/lb (−259°F, 14.7 psia, liquid)
Then by use of Equation 13.6
Effect of heat leak on production—Use the same cycle but now have a 5°F
temperature difference at the warm end of the heat exchanger (t
1
, t
3
) and a heat
leak, 15 Btu/lb. This change lowers the recycle gas outlet temperature to 75°F
and h
3
390 Btu/lb (75°F, 14.7 psia)
The liquid fraction generated now becomes

Effect of pressure on production—Determine how liquid production is affected
by increasing the pressure on the inlet gas to 2,000 psia from 1,500 psia.
h
1
337 Btu/lb (80°F, 2,000 psia)
This example illustrates the effect of pressure and heat exchanger perfor-
mance on liquid yield. For example, if the warm end ∆T for the heat exchanger
is approximately 77ºF (43ºC), a very unlikely value, the liquid yield is reduced
to zero, even if no external heat leaks are present.
This outcome raises the question of whether an optimum pressure exists. In
Equation 13.6, f will be a maximum when is a maximum because
the other terms are independent of inlet pressure. The enthalpy of the liquid, h
2
,
depends only on the liquid receiver pressure, which we hold constant at the lowest
pressure (approximately 14.7 psia [1 bar]). Also is independent of pressure
∆T
f
h h q
h h
L

− −


− −


3 1
3 2
392 350 0
392 0
0 107 .
q
L
f
h h q
h h
L

− −


− −


3 1
3 2
392 350 15
392 0
0 069 .
f
h h q
h h
L

− −


− −


3 1
3 2
392 337 0
392 0
0 140 .
( ) h h q
L 3 1
− −
q
L
saturation table (Appendix B) the following values are obtained:
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 279
and is fixed at the lowest pressure and the highest temperature (the inlet
temperature for zero at the warm end of the heat exchanger). Thus we
maximize f when h
1
is a minimum. The mathematical criterion is
. (13.7)
Because thermodynamic optimum pressure will occur when
or when the inlet conditions are on the inversion curve. However, many other factors
must be considered in selecting the economically optimum inlet conditions.
Considerable improvement can be achieved in this simple Joule-Thomson
cycle, but at the expense of added equipment and complexity of operation. The
addition of an external source of refrigeration markedly improves efficiencies,
as does the use of a double expansion of the high-pressure gas instead of a
single expansion. Although both of these techniques are extensively used in air
liquefaction plants, only the dual-expansion process has found favor in LNG
processing.
Joule-Thomson cycle in a Richmond, B.C. plant, near Vancouver, B.C. This
plant served a stranded utility, and its total production was transported over-
land by truck. It was designed and built to allow easy movement to a new
location.
Feed to the plant is obtained from a natural gas pipeline at 40°F (4°C) and
pressures in excess of 300 psig (20 barg). The inlet gas is regulated to 300 psig
(20 barg) and passed through a molecular sieve dryer to remove both water
vapor and carbon dioxide. The gas then is compressed to 3,000 psig (210 barg)
in an electrically-driven, two-stage reciprocating compressor. After passing
through the three-stream heat exchanger, the gas undergoes a double Joule-
Thomson expansion, first to 300 psig (21 barg), and then to 10 psig (0.7 barg)
to liquefy the stream. The LNG is transferred to one of the two storage tanks
at the facility, either a 21,000 gallon (Imperial) horizontal cylindrical tank that
uses vacuum perlite insulation or a 35,000 gallon (Imperial) aluminum tank
was shipped to Squamish, British Columbia, by truck in a 5,000-gallon (Impe-
rial) trailer. At Squamish, the LNG is transferred to a 21,000-gallon (Imperial)
storage tank before regasification and distribution in the town’s natural gas
system.
Unfortunately, no specific details are available regarding the composition of
the natural gas or the specific economics of the facility. Truck transport was
apparently more economical than pipelining the gas between the two locations,
a distance of about 40 miles. In a relatively small installation such as the Richmond
facility, with a liquefaction capacity of 140 Mscfd (3.96 MSm
3
d), the low capital
cost and simplicity of design in the Joule-Thomson cycle counterbalanced the
inherent thermodynamic inefficiency, and thus made this cycle a logical choice.
h
3
∆T


h
P
T
1
0
|
.

`
}

µ



1
C
H
P T
P
( ) , µ 0
Figure 13.8 shows the schematic of a commercial facility that used the
embedded in the ground. (See Section 13.4 for storage tank details.) The LNG
© 2006 by Taylor and Francis Group, LLC
280 Fundamentals of Natural Gas Processing
For larger installations, however, the increase in thermodynamic efficiency avail-
determining factor in their selection.
13.3.2 EXPANDER CYCLES
The point was made during the discussion of the Joule-Thomson expansion that
it was a thermodynamically irreversible process. Expansion of high-pressure gas
to the lower pressure in a reversible or nearly reversible manner provides two
distinct improvements over the Joule-Thomson expansion. First, in the reversible
expansion, a large fraction of the work required to compress the gas can be
recovered and used elsewhere in the cycle. This property provides an increase in
cycle efficiency. Second, the reversible process will result in a much larger cooling
FIGURE 13.8 Joule-Thomson liquefaction plant (Blakely, 1968).
LNG storage
LNG
300 psig
−160°F
LNG
10 psig
−260°F
300 psig
3000 psig 300 psig
Dryer
Natural gas pipeline, >300 psig, 40°F
J-T valve
J-T valve
able in both the expander and cascade cycles (See section 13.3.3) becomes the
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 281
effect. For example, reversibly and adiabatically expanding methane gas from 75
psia and 80°F (5.1 bar, 27°C) to 14.7 psia (1.01 bar) cools the gas to −94°F
(−70°C), a temperature drop of 174°F (97°C). A Joule-Thomson expansion
between the same pressure limits cools the gas approximately 4°F (2.2°C). As
possible over a given pressure range.
Several options are available for selection of expanders for LNG use, both
in the type of expander and in the basic cycle itself. Expanders are basically
compressors with the flow reversed and, as with compressors, positive displacement
and dynamic expanders are available. In 1902, Georges Claude pioneered
expander use in air liquefaction. Claude’s expander was a reciprocating machine,
as were most early machines used in cryogenic processes, such as those developed
by Heylandt in 1912 and later by Collins (1947). Barron (1966) reports reciprocating-
machine adiabatic efficiencies of 70 to 80%. He attributes reciprocating expander
inefficiencies to four causes:
• Inlet and outlet valve losses
• Incomplete expansion
• Heat transfer
• Piston friction
Reciprocating machines are rarely used in LNG facilities.
Similar to dynamic compressors, dynamic expanders can be centripetal flow
or axial flow. In centripetal machines (i.e., turboexpanders), the gas enters
through nozzles around the periphery of the wheel, expands, and transmits work
to the wheel, which causes it to rotate, and finally exhausts at low pressure at
the axis of the machine. Chapter 7 provides more details. Axial-flow expanders
have as their counterparts steam turbines. Centripetal machines have isentropic
efficiencies on the order of 85 to 90%, whereas axial-flow expanders are about
80% efficient (Swearingen, 1968). Turboexpanders are high-speed machines,
generally designed to operate from 10,000 to 100,000 rpm, depending on the
throughput.
For design purposes, several techniques may be used to compute the expected
enthalpy change, but the simplest and apparently satisfactory method is to use
the ideal value from a P-H or T-S diagram, and correct this value with the
anticipated turboexpander efficiency (Swearingen, 1968; Williams, 1970). The
work generated in the expander must be removed from the system if the full
thermodynamic efficiency of the cycle is to be realized. The general practice in
large-scale operations is to couple the turboexpander to a gas compressor. Recip-
rocating expanders would naturally be coupled with reciprocating compressors,
and turboexpanders coupled with centrifugal compressors. The available expander
work can be very large. Swearingen (1968) states that a turboexpander handling
500 MMcfd (14 Sm
3
/d) at pipeline pressure would develop 10,000 hp (7,500
kW). Surprisingly, the turbine rotor would only be 18 inches in diameter. In
small-scale operations, recovery of the expander work is often not economically
noted in Chapter 7, adiabatic reversible turboexpansion provides the most cooling
© 2006 by Taylor and Francis Group, LLC
282 Fundamentals of Natural Gas Processing
feasible. In this case, the turboexpander is simply coupled to a braking device
that dissipates the work. Expander−compressor combinations require consider-
able care in their selection and operation. Swearingen (1970) and the Engineering
Data Book (2005b) discuss what must be considered in the selection, operation,
and maintenance of turboexpanders.
As mentioned previously, several options are available in the type of expander
cycle. All expander cycles fall into two groups: closed cycles and open cycles.
Note that most expander cycles have J-T valves as well as turboexpanders.
13.3.2.1 Closed Cycles
In a closed expander cycle, the fluid being expanded is not the fluid to be liquefied;
the expander simply acts as an external source of refrigeration, similar to the
nitrogen may be used in a closed expander system to liquefy natural gas. A very
nitrogen is expanded, and the cold gas is then used to cool and liquefy the natural
gas stream. Actual cycles for producing LNG are far more complex. Hathaway and
Lofredo (1971) provide a complete process flow sheet for a plant that has four
warm heat exchangers, one large nitrogen compressor, and three turboex-
pander/compressor combinations.
The closed cycle has several advantages over the open cycle, in which the
natural gas itself is expanded. First, if nitrogen is used, safety is enhanced, because
the closed cycle reduces the number of processing steps in which flammable natural
gas is used. Second, the closed nitrogen cycle has been reported (Anonymous,
1970) to require simpler and less expensive shutdown procedures than its open
cycle counterpart and appears to be the most economical process under many
conditions. Finally, because the natural gas is not passing through the expander,
the process purification system is not so critical. Gas passing through the high-
speed expander must be free of condensed phases and any components that solidify
at the expander exhaust temperature, because deposition on the rotor will destroy it.
13.3.2.2 Open Cycles
An open expander cycle uses the gas being liquefied as the expanding fluid and
has the advantage over the closed cycle of being less complex. A basic expander
the expander is simply used as a source of refrigeration, and the high-pressure
gas is liquefied as it expands through the Joule-Thomson valve. The first law of
thermodynamics for a steady-state flow system applied to the two heat exchang-
ers, the expander, and the liquid receiver gives the following equation:
, (13.8) f
h h q
h h
e h h
h h
L

− −

+


3 1
3 2
4 6
3 2
( )
simple schematic of a closed cycle is shown in Figure 13.9. The compressed
cycle is shown in Figure 13.10. In this example, the cold exhaust stream from
propane refrigeration discussed in Chapter 7. For example, in LNG production,
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 283
where e is the fraction of the gas going to the expander and (h
4
− h
6
) is the work
the mass flow rate into the liquefier, the mass flow rate of liquefied product, and
the mass flow rate to the expander, respectively. A more detailed analysis of the
overall system is possible (Dodge, 1944). The cycle shown is only one of a
number of possible process arrangements.
is the Chula Vista plant of the San Diego Gas and Electric Company (Hale, 1966).
The plant receives 25 MMscfd (0.71 Mm
3
/d) from a natural gas pipeline at 300
psia (21 bar) and 90°F (32°C). The gas is first prepared for liquefaction by removal
of CO
2
, H
2
S, and water by physical adsorption on a molecular sieve. The stream
then splits, with about 21 MMscfd (0.59 Mm
3
/d) going to the expander to provide
refrigeration. The refrigerant is initially cooled in the first heat exchanger before
going to a separator. Liquid from the separator expands through a Joule-Thomson
valve. It recombines with the vapor from the separator, which has been through
a turboexpander where the pressure drops to 60 psia (4.1 bar) and the temperature
drops to −175°F (−115°C), and through the second heat exchanger. This stream
provides cooling to both the incoming refrigerant stream and the fraction to be
liquefied in the first exchanger. The gas then is compressed to 82 psia (5.6 bar)
before being odorized and sent to the local power plant. Compression comes from
work done by the turboexpander.
The 4 MMscfd (113 MSm
3
/d) of gas in the liquefier stream passes through
all three heat exchangers and a Joule-Thomson expansion valve. Liquid and vapor
are then separated; the vapor stream passes through the heat exchangers and then
goes to fuel for the power plant. Three-fourths of the gas that enters the liquefier
becomes liquid.
FIGURE 13.9 Simple closed cycle liquefaction process.
Compressor Expander LNG product
Treated natural
gas feed

f

e
done by the expander. The quantities m, m , and m in Figure 13.10 represent
An industrial LNG facility that uses an open expander cycle (Figure 13.11)
© 2006 by Taylor and Francis Group, LLC
284 Fundamentals of Natural Gas Processing
When desired, the LNG is gasified by pumping the liquid to 460 psia (32
bar) and vaporizing it in a hot water heat exchanger. The gas, at 400 psia (27.5
bar) and 60°F (16°C), is then ready for distribution. The vaporizer capacity is 60
MMscfd (1.7 MMSm
3
/d).
The LNG is stored in a single 175,000-barrel (27,800 m
3
) aboveground
storage tank but has 1 MMscfd (28 MSm
3
/d) of boil-off. The boil-off provides
some refrigeration and is compressed and combined with the vapor from the
separator before going to the power plant. With a net liquefaction rate of 2 MMscfd
(57 MSm
3
/d), 315 days are required to fill the storage tank, but only 10.3 days
are required to empty the tank if vaporization is at the maximum rate. This
outcome matches the gas demand, as the company typically has surplus gas
available about 300 days a year. During this period, the storage tank is filled.
During the much shorter periods of peak demand, the LNG is vaporized and
placed in the distribution system.
FIGURE 13.10 Open cycle expander plant.
Liquid
product
J-T valve
Turboexpander
1
m
7
5
m
f
4
10
8
2
6
m
e
Purified gas feed

e
r
m
o
d
y
n
a
m
i
c

b
o
u
n
d
a
r
y
q
L
.
.
.
3
9
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 285
13.3.3 CASCADE CYCLES
One source of thermodynamic inefficiency is the finite temperature difference
that must be present for heat transfer. The maximum thermodynamic efficiency
in a liquefaction cycle is realized when the heating curve of the refrigerant
corresponds to the cooling curve of the natural gas being liquefied. This relation
means that both ∆T (and thus ∆S) are zero for the heat transfer process (Dodge,
and heating curves for a mixed refrigerant system and a three-fluid classical
cascade system. The natural gas and mixed refrigerant curves show marked
curvature because the fluids are mixtures. Thermodynamically, the mixed refrigerant
FIGURE 13.11 Peak-shaving plant with open cycle.
J-T
valve
J-T
valve
Boil-off
compressor
Compressor-expander
Vaporizer
To
pipeline
Odorant
To
power
plant
Odorant
Boil-off
compressor
Recycle
compressor
Mol sieve
cleanup
LNG
pumps
Pipeline gas
1944). Figure 13.12 is a schematic of a cooling curve for a natural gas system
© 2006 by Taylor and Francis Group, LLC
286 Fundamentals of Natural Gas Processing
comes closest to a reversible process because it minimizes the temperature dif-
ference between the two fluids.
The classical cascade attempts to approximate the cooling curve by use of a series
of refrigerants (usually three) in separate loops. Use of more than three refrigerants
allows a closer approximation to the cooling curve but with the penalty of extra
equipment and added cycle complexity. Both mixed refrigerant cascades and classical
cascades will be discussed in more detail in subsequent sections.
The classical cascade was one of the earliest cycles developed and used in
the liquefaction of the so-called permanent gases, helium and hydrogen, and was
used both by Pictet in 1877 and by Kamerlingh-Onnes in 1892−94 for the
liquefaction of air. The method was also used in the first LNG plant built in
the United States at Cleveland, Ohio, in the 1940s. The classical cascade is
attractive because it can be very efficient thermodynamically. Recently, the
classical cascade has been modified with the introduction of the mixed refrig-
erant cascade. Because both types of cycles are in use for LNG production, we
discuss the more important features of both processes.
13.3.3.1 Classical Cascade
The classical cascade process starts with a vapor that can be liquefied at ambient
temperature by the application of pressure only. The liquid formed by pressur-
ization is then expanded to a lower pressure, which results in a partial vaporization
FIGURE 13.12 Cooling curve for natural gas (

) and the corresponding warming curves
for the mixed refrigerant (- - -) and cascade (
. . . .
) cycles.
0 20 40 60 80 100
+40
−20
−80
−140
−200
−260
T
e
m
p
e
r
a
t
u
r
e

°
F

Percent enthalpy removed
Natural gas
Mixed refrigerant
Cascade
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 287
and cooling of the remaining liquid. This cold liquid bath is then used to cool a
second gas so that it may also be liquefied by the application of moderate pressure
and then expanded to a lower pressure. The temperature reached in the expansion
of the second liquid will be substantially lower than that achieved by the expan-
sion of the first liquid. In principle, any number of different fluids may be used
and any desired temperature level can be reached by use of the appropriate number
of expansion stages. In practice, however, three fluid and three levels of expansion
are normal. In the liquefaction of natural gas, the custom now is to use a propane-
ethylene (or ethane)-methane cascade, although the first LNG plant at Cleveland,
Ohio, used an ammonia-ethylene-methane cascade (Miller and Clark, 1941).
A simplified process flow diagram for the first stage in a cascade cycle is
faction techniques, in that the cooling is obtained principally from external cir-
cuits, with one final Joule-Thomson expansion for liquefaction. The Joule-Thomson
expansion and many turboexpander processes use the gas to be processed as their
refrigerating medium.
Figure 13.13 is a simplified schematic of a hypothetical two-fluid cascade
designed to produce a liquid product at a low temperature. The process shown uses
variations of a two-stage and three-stage refrigeration system (see Chapter 7), with
two different working fluids. As noted above, plants that use this cycle normally
use three fluids, but we discuss a two-fluid cycle to keep the schematic simple.
Following the flow from the compressor discharge (1) for the high-temperature
working fluid, the pure fluid is cooled in heat exchanger E-1, expanded through a
Joule-Thomson valve (2), and goes to liquid-vapor receiver R-1 (3). The liquid in the
FIGURE 13.13 Schematic of a hypothetical two-fluid cascade cycle.
Natural gas
feed
Liquid-vapor
product
1
2
3
4
5
6
7
8
9
10
E-1 E-4
E-2
E-5
E-6
R-1 R-2 R-3 R-4 R-5
E-3
shown in Figure 7.2, and the operation of the refrigeration system is discussed
in Chapter 7. The cascade process differs from the previously described lique-
© 2006 by Taylor and Francis Group, LLC
288 Fundamentals of Natural Gas Processing
receiver cools the gas feed, and exiting vapor flows back through heat exchanger E-
1 to the second-stage compressor suction (4). In addition to cooling the gas feed, part
of the liquid from the receiver (3) flows through heat exchanger E-2, where it is cooled
before expansion through a Joule-Thomson valve (5), which creates a liquid−vapor
mixture in the receiver R-2 (6). Both the temperature and the pressure of the liquid−
vapor mixture in receiver R-2 are lower than in receiver R-1. The liquid in receiver
R-2 further cools the gas feed. The vapor passes through a liquid separator, two heat
exchangers E-1 and E-2, and goes to the compressor suction (7). Liquid from receiver
R-2 goes to heat exchanger E-3 (8) to cool the pure fluid in the second cascade before
expansion through a Joule-Thomson valve (9) to create a cooler liquid−vapor mixture
in receiver R-3 (10). Essentially the same process is repeated for the second working
fluid, which goes through three stages of refrigeration. The feed gas progressively
cools until it leaves receiver R-5. It then passes through the final Joule-Thomson
valve, and becomes a liquid−vapor mixture at the final desired temperature and
pressure. The liquid goes to storage, while the cold vapor is used for heat exchange
before being recycled or used for fuel. For simplicity this loop is not shown.
Although the figure and its discussion are simplified, they do illustrate the
fundamental features of a classical cascade. Industrial cascades for LNG produc-
tion use three pure fluids with two or three stages of refrigeration for each fluid.
Andress (1996) discusses an industrial application of the classical cascade. Note
that for the classical cascade system to function, the vapor pressure curves of the
pure fluids must overlap, as shown in Figure 13.14.
13.3.3.2 Mixed-Refrigerant Cascade
As mentioned earlier, the maximum thermodynamic efficiency in a liquefaction
cycle is realized when the heating curve of the refrigerant corresponds to the
cooling curve of the natural gas being liquefied. The classical cascade attempts
FIGURE 13.14 Qualitative plot of vapor pressure for two pure refrigerant fluids that
shows the overlapping region needed for cascade cycles.
Critical point
P
r
e
s
s
u
r
e
Temperature
Overlap
V
a
p
o
r

p
r
e
s
s
u
r
e

o
f


u
i
d

A
V
a
p
o
r

p
r
e
s
s
u
r
e

o
f


u
i
d

B

© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 289
to approximate the cooling curve by use of three pure refrigerants in separate
loops. However, replacement of several pure refrigerants in separate cycles with
a single refrigerant composed of many components condensing at various tem-
peratures in one cycle makes possible a more closely matched cooling curve of
the natural gas. In addition, only a single compressor is required for the refrigerant.
This feature is the principle for the mixed-refrigerant cascade, the MRC (Linnett
and Smith, 1970), the autorefrigerated cascade, the ARC (Salama and Eyre, 1967),
and the one-flow cascade OFC process (Forg and Etzbach, 1970).
The technical feasibility of mixed-refrigerant cycles was apparently first
proved experimentally by Kleemenko (Forg and Etzbach, 1970) in 1959. Actual
cycles that used mixed refrigerants appear in a variety of configurations, but they
can be loosely grouped into two basic categories: closed cycles and open cycles.
(single-flow mixed-refrigerant). The refrigerant would probably be a mixture of
nitrogen, methane, ethane, propane, butane, and perhaps pentane; the exact com-
position depends upon the composition of the natural gas being liquefied. The
refrigerant mixture is compressed and then partially condensed in a water-cooled
exchanger. The refrigerant then undergoes a series of pressure reductions and
liquid-vapor separations to provide the cold fluid needed in the heat exchangers to
liquefy the natural gas. The temperatures attained in the various heat exchangers
depend on the composition of the refrigerant and the pressure to which the gas
is initially compressed. These operating parameters are selected to approximate
the cooling curve of the natural gas being liquefied. In this cycle, the natural gas
passes through all four heat exchangers in series and is then expanded into a
industrial example of this type system.
system, the propane precooled, mixed refrigerant system. It is the most commonly
used cycle for LNG baseload plants. The process uses both external propane
refrigeration and refrigeration from expanding the mixed refrigerant. A three-
[2005c] for propane refrigeration systems) provides the initial cooling for both
the natural gas and the mixed refrigerant in exchangers HX1, HX2, and HX3.
Compressor C1 provides the work for the propane refrigeration. Further cooling
is supplied by the expansion of the mixed refrigerant liquid and vapor leaving
the cold separator S1. Both of these streams and the process gas are chilled by
the expanded mixed refrigerant in HX4 and HX5. The low-pressure mixed refrig-
erant is recompressed by use of the two-stage mixed-refrigerant compressor C2.
Perez et al. (1998) provide a complete description of a large propane precooled,
mixed-refrigerant system.
the natural gas stream to be liquefied is physically mixed with the refrigeration
cycle stream. This mixing can take place before, during, or after the compression
process, depending upon the pressure at which the natural gas feed is available.
After compression, the united gas streams are partially condensed in a water-cooled
Closed Cycles—Figure 13.15 shows a schematic of a simple closed cycle
separator, where the liquid and vapor fractions are separated. Figure 13.16 is an
Figure 13.17 is a schematic of the popular closed cycle mixed refrigerant
Open Cycle—In the open cycle system shown in simplified form in Figure 13.18,
stage propane refrigeration cycle (see Chapter 7 and Engineering Data Book
© 2006 by Taylor and Francis Group, LLC
290 Fundamentals of Natural Gas Processing
FIGURE 13.15 Schematic of simple closed-cycle, single-flow mixed refrigerant (Forg
and Etzbach, 1970).
FIGURE 13.16 Schematic of commercial closed-cycle PRICO® system. (Adapted from
Price et al. 2000.)
Natural gas feed
Fuel
LNG
Interstage
separator
Compressor
Suction
drum
Condenser
Refrigerant
separator
Natural gas
LNG
Heavy
hydrocarbons
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 291
or air-cooled heat exchanger, and then separated into liquid and vapor fractions
in a separator. From this point, the process is similar to the closed cycle system.
That is, the liquid fractions are expanded, which results in vaporization and
cooling, and these cold streams are used as the coolant in the heat exchangers.
The vapor from the last separator is condensed in the final heat exchanger, and
then expanded and separated into an LNG product and a flash gas that would
generally be used for plant fuel. To prevent heavy hydrocarbons from plugging
in the low-temperature portion of the cycle a liquid slipstream may be withdrawn
at a relatively high temperature.
Summary—The mixed refrigerant cycles discussed above possess several
advantages over the classical cascade system. The principal advantage is the use
of a single-compressor refrigerant system (excluding the propane precooling) in
FIGURE 13.17 Schematic of a closed-cycle propane precooled mixed refrigerant system.
Shaded areas and dotted lines represent propane cooling system. (Adapted from Foerg et
al., 1998.)
LNG
Treated natural gas
HX1
HX2
HX3
HX4
HX5
C1
C2
S1
© 2006 by Taylor and Francis Group, LLC
292 Fundamentals of Natural Gas Processing
place of the three refrigerant compressors and cycles of the standard cascade.
This configuration provides simplification in instrumentation and piping and a
better use of compression power. Further advantages include the ability to readily
change the composition of the refrigerant for cycle optimization, should the
composition of the feed gas change, and the ability to extract cycle refrigerants
directly from the feed gas. The disadvantage when compared with the standard
cascade is the necessity for having facilities to recover, store, and blend the
components in the refrigerant cycle.
A number of studies compare the various cycles for a particular applica-
tion. These studies include Vink and Nagelvoort (1998), Foerg et al. (1998),
Linnett and Smith (1970), Finn et al. (2000), Foglietta (2002), and Kotzot
(2003).
13.4 STORAGE OF LNG
Discussions of LNG storage facilities are normally divided into two major
categories: aboveground and in ground. Each has three natural subcategories, and
each is discussed below. Also, LNG storage involves a feature peculiar to the
FIGURE 13.18 Open cycle schematic. Refrigerant and process stream are mixed (Forg
and Etzbach, 1970).
Natural gas
feed
Fuel
LNG
Liquid
Slipstream
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 293
storage of cryogenic liquid mixtures, stratification. This section includes a dis-
cussion of this important phenomenon.
13.4.1 CRYOGENIC ABOVEGROUND STORAGE
Three basic types of aboveground storage vessels are in use:
• Steel
• Prestressed concrete
• Hybrid (combinations of steel and concrete)
Each type of construction will be discussed briefly.
13.4.1.1 Steel
double-, and full-containment steel tanks. Storage of LNG in aboveground metal
tanks is a widely accepted method for both baseload and peak shaving uses. Nine
percent−nickel steel is the most widely used metal for large LNG tanks. Because
of its high cost, stainless steel is generally used only for small vessels, LNG plant
piping, and heat exchangers. The conventional configuration is a double-walled,
flat-bottomed tank, with the annular space between the walls filled with an insulating
material. Long (1998) discusses two types of 9% nickel steel construction, the
single-containment and the double-containment tank. The single-containment tank
is obviously less expensive to build but has the obvious disadvantage of only one
containment wall that is compatible with LNG. An economic study by Long (1998)
indicated a 12 to 17% cost differential between the full- and double-containment
FIGURE 13.19 Single-containment tank. (Adapted from Long, 1998 and Kotzot, 2003.)
Carbon steel
outer tank
Perlite
insulation
Reinforced concrete pad
Electrical heating
elements
Suspended deck
Carbon steel
roof
9% nickel steel
inner tank
Base
insulation
Figures 13.19, 13.20, and 13.21 show three typical configurations for single-,
© 2006 by Taylor and Francis Group, LLC
294 Fundamentals of Natural Gas Processing
tanks, but his analysis did not include the additional cost of secondary impoundment
or additional fire protection necessary for the single-containment design. Conse-
quently, double-containment tanks may actually be the most economical choice. In
a full-containment tank, the roof is also constructed of concrete.
13.4.1.2 Concrete
Hundreds of prestressed concrete tanks and reservoirs have been built for many
uses, including the storage of liquid oxygen (Closner, 1968), which is both heavier
and colder than LNG. These tanks have been in continuous use since 1968.
FIGURE 13.20 Double-containment tank. (Adapted from Long, 1998 and Kotzot, 2003.)
FIGURE 13.21 Full-containment tank. (Adapted from Long, 1998 and Kotzot, 2003.)
Carbon steel
outer tank
Perlite
insulation
Reinforced concrete pad
Electrical heating
elements
Suspended
deck
Carbon steel
roof
9% nickel steel
inner tank
Base
insulation
Concrete
outer
shell
Carbon steel
outer tank
Perlite
insulation
Reinforced concrete pad
Electrical heating
elements
Suspended deck
Concrete roof
9% nickel steel
inner tank
Base
insulation
Concrete
outer
shell
Concrete outer
shell
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 295
Prestressed concrete tanks may be constructed at grade, below ground, or partially
below ground, depending on site conditions or other factors.
Figure 13.22 shows typical installation of a prestressed concrete storage tank.
The tank was constructed with its base at grade but was completely surrounded by
an earthen berm. The tank had a capacity of approximately 2 billion cubic feet
of gas (approximately 600,000 barrels [95,000 m
3
]) and was built for a peak
shaving facility on Staten Island, New York. Prater (1970) provides details of
the tank construction. The reinforced concrete walls were insulated on the inside
of the tank with polyurethane. The LNG was isolated from the insulation with
a thin laminate film that had additional protective insulation. The anticipated
boil-off was 0.06% per day of the tank volume. Two centrifugal blowers, each
capable of handling 2 MMcfd (60 MS m
3
/d) at atmospheric pressure and −200°F
(−129°C) suction, were used to compress the boil-off vapors to 10 psig (0.7 barg),
and the boil-off was subsequently reliquefied. The normal operating tank pressure
was 4 inches (10 cm) of water, and a system of diaphragm-operated valves was used
to add or release gas rapidly in the event of abnormal atmospheric conditions, such
as those caused by a hurricane. The tank was destroyed in an unfortunate maintenance
construction accident in 1973. During repairs to the inside of the storage tank, a fire
started. The increase in pressure that resulted was rapid and lifted the concrete roof,
which then collapsed and killed 37 workers inside the tank (Federal Energy Regula-
tory Commission, 2005).
Concrete tanks may be constructed where the concrete is protected from direct
contact with the LNG by a membrane* or where the LNG directly contacts the
concrete. The Portland Cement Association (Monfore and Lentz, 1962) showed
that concrete cured at ordinary temperatures can be used to form tanks for holding
cryogenic fluids. The mechanical properties of concrete are not significantly
FIGURE 13.22 Prestressed concrete storage tank for LNG (Duffy et al., 1967).
* Membranes here denote sheets of impervious material, typically metal, that prevents the LNG and
Earth berm
Concrete
Select fill
Ballast
Insulated
roof
Insulation
and liner
Spring-loaded
trusses
gas vapors from contacting the tank walls, not membranes for separation as discussed in Chapters 5,
6, and 7.
© 2006 by Taylor and Francis Group, LLC
296 Fundamentals of Natural Gas Processing
impaired at low temperatures. The compressive strength of concrete at −250 to
−150°F (−156 to −101°C) is almost triple that of concrete at room temperature.
Legatos et al. (1995) and Jackson and Powell (2001) provide details on the
construction and economics of full-containment prestressed-concrete tanks. Jackson
and Powell (2001) include data on the rate of LNG permeation of the concrete.
13.4.1.3 Hybrid Construction
Storage tanks constructed using both prestressed concrete and steel are popular.
Nishizaki et al. (2001) discuss the construction of a 180,000-m
3
(1.2 MMBbl) LNG
storage tank that is claimed to be the world’s largest aboveground tank. The
structure consists of an inner tank of 9% nickel steel, an outer prestressed-concrete
tank, with a cryogenic insulation between the two tanks. Rapallini and Bertha
(1998) discuss the conversion of two single-containment 9% nickel steel tanks
into double-containment tanks by the addition of an outer concrete tank. Giribone
and Claude (1995) describe a novel concept for an aboveground tank that features
a 1.2-mm thick stainless-steel membrane combined with a prestressed concrete
outer wall. In this configuration, the membrane acts to contain the LNG, but all
the hydrostatic load is taken by the concrete.
13.4.2 CRYOGENIC IN GROUND STORAGE
Three basic types of in ground storage have been used:
• Conventional concrete or steel tanks in an underground configuration
• Tanks formed around a frozen-earth cavity
• Mined caverns
All three types will be discussed briefly.
13.4.2.1 Conventional Tankage
With aboveground tank storage discussed in the previous section, the walls must
supply all of the mechanical strength. In ground tanks may use either the sur-
rounding earth to provide mechanical support or an in-pit construction in which
the tank is built as a separate unit and the pit provides containment in case of
leakage or rupture.
As an example of the first type, two of the world’s largest LNG tanks are
200,000-m
3
(1.3 MMBbl) in ground storage units installed at the Ohgishima LNG
terminal of the Tokyo Gas Company (Umemura et al., 1998). The tanks are unique
not only because of their size but also because of the fact that the entire tank,
including the domed roof, is buried. Each of the cylindrical tanks is constructed
entirely of reinforced concrete and has sidewalls of 2.2 m (7.2 ft) thickness, with
a 9.8 m (32 ft) thick bottom slab, 78 m (255 ft) in diameter, designed to withstand
ground water uplift. The domed roofs are also of reinforced concrete (1 m
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 297
thickness at center and 2.5 m at circumference). Because the tanks are totally
buried, each roof supports its own weight (15,000 metric tons) plus the weight
of soil covering it (40,000 metric tons). The inside of the tanks, including the
domes, are lined with cryogenic insulation and 2-mm stainless-steel membranes
for liquid and gas tightness. Heating systems surround the sides and bottoms to
control ground freezing.
Okamoto and Tamura (1995) discuss the first in-pit storage unit built in Japan,
an 85,000-m
3
(0.53 MMBbl) unit. In their design, they chose a storage tank
identical to an aboveground, cylindrical and vertical, metallic double-shell, flat-
bottom tank, and installed the tank inside an in ground reinforced-concrete pit.
Dragatakis and Bomhard (1992) discuss in some detail the problem of tank
construction in areas prone to severe earthquakes, and they discuss in-pit units
at the Revythousa Island LNG facility, located 30 km from Athens, Greece. In
their in-pit approach, the tank is freestanding in the pit and functions like an
aboveground tank during a seismic event.
13.4.2.2 Frozen-Earth Cavities
Figure 13.23 shows a sketch of a frozen-earth storage container. The cavity is
initially cooled by spraying LNG into the vapor space. The roof reaches its steady-
state temperature rapidly. Because of the low thermal conductivity of the frozen
earth, the surrounding soil may take several years to reach its steady-state temperature.
FIGURE 13.23 Sketch of frozen-earth LNG storage container.
Spray nozzles
9% nickel steel
roof
Freeze pipe
Insulation
LNG
© 2006 by Taylor and Francis Group, LLC
298 Fundamentals of Natural Gas Processing
The final effective thermal conductivity attained by the frozen earth depends
strongly on the type of geologic formation and the moisture content of the earth.
Four early LNG plants incorporated frozen-earth cavities for storage. The
first cavity was located in Arzew, Algeria; the next was connected with a peak
shaving facility near Carlstadt, New Jersey. Two cavities were then constructed
at the Tennessee Gas Pipeline Company’s peak shaving facility near Hopkinton,
Massachusetts. Finally, four frozen-ground storage units were built at the Canvey
Island receiving facilities in England.
Of the four plants mentioned that incorporated frozen-earth cavities, in 1983,
only the cavities at Arzew, Algeria were in service (Boulanger and Luyten, 1983).
The two plants in the United States were abandoned in favor of other forms of
storage because of the failure of the frozen-earth cavities to perform satisfactorily
in peak shaving applications. The English facility is an offload site. Because of
the lack of published operating data, assessment of differences in performance
among these four sites and determination of the reasons for abandoning some of
the cavities is not possible.
13.4.2.3 Mined Caverns
In this storage concept, a subterranean cavity is created to hold the LNG, with
the cavity walls either in direct contact with the liquid or separated by an insu-
lating wall. Boulanger and Luyten (1983) describe test results in a frozen-clay
cavern at Schelle, Belgium. They showed that the horizontal cavern, 3 m (10 ft)
in diameter and 30 m (100 ft) long, located 23 m (75 ft) beneath the surface in
a layer of boom clay, held liquid nitrogen for at least 10 weeks.
Ahlin and Lindblom (1992) discuss a cavern model in which the LNG is stored
in vertical rock caverns, with the LNG separated from the cavern by a layer of
concrete cast against the wall and an inner wall of Invar steel on insulating panels.
Presently, neither of the above techniques has been applied commercially.
13.4.3 ROLLOVER
In 1971, at the LNG terminal in La Spezia, Italy a sudden increase in pressure
occurred, with subsequent substantial venting of an LNG tank that had received
delivery some 18 hours earlier. Fortunately, no damage or injuries occurred. This
incident was the first of several similar instances of large, sudden vapor releases
that became known as “rollover” (Action and van Meerbeke, 1986). This section
provides a short overview of the cause of the releases and how they are prevented.
13.4.3.1 Cause of Rollover
Storage of cryogenic liquids is unique in that the heat leak continually warms
the liquid, whereas fluids stored at ambient temperature experience both heating
and cooling. To dissipate the heat influx, the cryogenic liquid boils off. For pure
liquids, such as liquid nitrogen, this process causes no problems because the
liquid is pure and no composition change takes place.
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 299
points differing by hundreds of degrees. Of these liquids, only nitrogen and
methane vaporize because they are the only components with a significant vapor
pressure at LNG temperatures.
First, consider a tank that contains LNG that is well mixed, with no nitrogen
present. As methane boils off, the LNG “weathers,” which increases the concen-
tration of the heavier components and, thus, increases the boiling point of the
liquid. An increase in density occurs because of the loss of methane, which offsets
the density decrease caused by temperature increase. Heat leak from the bottom
and sides of the tank causes continual mixing through natural convection. Over
time, this mixture slowly warms and increases in density.
Figure 13.24 shows a tank after the addition of a fresh batch of LNG on top
of the existing liquid, which creates two stratified layers. Initially, the top layer
is both lighter and cooler than the bottom layer. However, the top layer increases
in density as methane vaporizes. At the same time, the bottom layer continues
warming because of heat leak, but it cannot vaporize because of the hydrostatic
head of the top layer. This pressure lowers the bottom-layer density. (Although
mixing will occur between the layers at the interface, it is probably minimal,
considering the typically large volumes of liquid in each layer.)
Continual weathering in the top phase, which increases its density and, con-
currently, continual warming of bottom phase caused by heat leak, which lowers
its density, can lead to the bottom phase being the lighter phase. If the density
inversion becomes large enough to overcome the hydrostatic head, the potential
exists for the phases to flip, or rollover, which brings the warm liquid to the top,
with a sudden evolution of gas, because the hydrostatic head on the lighter phase
is gone. Tank size and type (aboveground, in ground, concrete, or metal) apparently
FIGURE 13.24 Schematic of tank that contains two stratified layers of LNG. Heat flux
into tank promotes mixing in each layer by natural convection.
Boil-off vapors
H
e
a
t
H
e
a
t
Vent
Fill lines
As Table 13.5 shows, LNG contains many components with normal boiling
© 2006 by Taylor and Francis Group, LLC
300 Fundamentals of Natural Gas Processing
have no significant affect on the occurrence of rollover (Acton and van Meerbeke,
1986). The slight changes in liquid composition, along with density difference
caused by temperature, are the important variables.
The heavy remnants (heel) in a tank are not a problem when lighter LNG is
added, because the density of the heel is sufficiently high to preclude a density
inversion. However, a thin, light layer on the top is the most prone to rollover,
although the evolution of gas may not be as severe as a thick layer because the
hydrostatic head is less.
Nitrogen at concentrations above 1 mol% in LNG mitigates the potential for
rollover when it is in the top layer because the nitrogen losses lower the density
instead of increasing it. However, if the nitrogen-rich layer is on the bottom, it
enhances the potential for rollover.
13.4.3.2 Prevention of Rollover
Rollover prevention requires knowledge of the temperature and density pattern
throughout the tank and the loading of new liquid accordingly. If the new charge
is lighter than the existing liquid, it goes in at the bottom of the tank. This
action forces immediate turnover of the liquids and provides the mixing
required, avoiding stratification. (Most tanks are far too large to use pumping
to turn over the liquid quickly, and pumping is kept to a minimum to reduce
additional liquid heating.) Conversely, heavier feed is filled at the top. Modern
tanks are instrumented with both density and temperature probes so that both
can be monitored for potential signs of stratification and the potential for
rollover.
To ensure that no danger is present, safety codes (e.g., NFPA-59A, 2006)
require that storage tanks have vent sizes many times larger than the normal boil-
off rate. A report by Groupe International des Importateurs de Gaz Natural
Liquefie (G.I.I.G.N.L, undated) found that of the 41 rollovers, only a few
exceeded 20 times the normal boil-off rate. European Standards require vent sizes
to handle as much as 100 times the normal boil-off rate to ensure that tanks will
not be overpressured.
The larger, offload storage facilities are less prone to rollover than peak
shaving facilities, because peak shaving liquid storage times can be much longer
than at the offload facility, allowing more time for “weathering and stratification.”
13.5 TRANSPORTATION
As noted in section 13.1 for large gas reserves often the only alternative to
pipelines is shipping by LNG. Three options are possible for transporting LNG:
• Truck transport
• LNG pipelines
• Marine carriers
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 301
Each method will be discussed, with the emphasis on marine carriers because
they are the most important.
13.5.1 TRUCK TRANSPORT
Cryogenic liquids, including liquid helium, liquid hydrogen, liquid nitrogen, and
liquid oxygen, are routinely moved by truck transport. Thus, over-the-road move-
ment of LNG is a relatively simple, straight-forward process that requires no new
technology. The major consumers of trucked LNG are vehicle fueling stations
and “stranded local utilities,” those who are not connected to the national network
of natural gas pipelines (Energy Information Administration, 2003a). They con-
sume a tiny faction of the LNG in the United States, 5 to 10 Bcf (140 MM to
280 MMSm
3
) equivalent per year (Energy Information Administration, 2004).
13.5.2 PIPELINES
The concept of long-distance LNG pipelines was studied by Hoover (1970), Katz
and Hashemi (1971), and Dimentberg (1971). Pumping liquid instead of com-
pressing gas makes the concept seem attractive. Hoover’s (1970) detailed study
of LNG pipeline concluded that en route refrigeration, primarily for removing
heat generated by friction losses, make pipelining economically feasible only in
certain situations, at distances less than 200 miles (320 km).
13.5.3 MARINE TRANSPORT
The first ship, Methane Pioneer, sailed from Lake Charles, Louisiana, to the
United Kingdom with 32,000 barrels (5,088 m
3
) of LNG on January 28, 1959.
Figure 13.25 shows that by the end of 2006 the LNG fleet will have 206 carriers
FIGURE 13.25 Size distribution of world’s LNG carrier fleet in 2003 (Energy Information
Administration, 2003b).
5
10
8
9
2
60
94
18
0
10
20
30
40
50
60
70
80
90
100
0 to 25 25 to
50
50 to
75
75 to
100
100 to
125
125 to
135
135 to
145
145+
Size of tanker, m
3
N
u
m
b
e
r
© 2006 by Taylor and Francis Group, LLC
302 Fundamentals of Natural Gas Processing
(Energy Information Administration, 2003b). Presently, the largest carriers in use
or under construction hold 165,000 m
3
(1 MMBbl), but ships that could hold
200,000 to 250,000 m
3
(1.3 to 1.6 MMBbl) are under consideration. With little
savings in economy of scale, reduced transportation costs are the prime motivation
for the larger capacities.
We briefly discuss several of the more important design considerations, and
then show details of some typical LNG carriers. Ffooks and Montagu (1967)
present a detailed discussion of the evolution of LNG ship design, and they list
several basic design criteria:
1. The low density of LNG and the requirement for separate water ballast
containment require a large hull, with low draft and high freeboard*.
2. The low temperature of LNG requires the use of special and expensive
alloys in tank construction. For free-standing tanks, only aluminum or
9% nickel steel are suitable, whereas for membrane tanks, stainless
steel or Invar is used.
3. The large thermal cycling possible in the storage tanks demands special
supporting arrangements for free standing tanks and membrane flexi-
bility in membrane designs.
4. The hull of the vessel is carbon steel, so good thermal insulation is
required between the tanks and the hull. In addition, for membrane tanks,
the insulation must be capable of supporting the full weight of the cargo.
5. The cargo handling equipment must be carefully designed to account
for thermal expansion and contraction.
Application of these principles in the design of LNG carriers resulted in
a number of different LNG containment concepts, but today only three sys-
tems are in general use, and they may be grouped into two designs, indepen-
dent tanks and membrane tanks, which use different membrane configurations
(Marshall, 2002). Both designs are presently in use, and as reported by Harper
(2002), approximately 130 ships are currently trading. More than half use
independent tanks, and the remainder use membrane tanks. Approximately
60 ships are on order; 40 use the membrane configuration and 20 use the
independent-tank design. Marshall (2002) has somewhat different figures: of
173 ships in service, 45% use independent tanks, 53% use membranes, and
2% use other configurations.
Presently, all LNG carriers are double-hulled. With two exceptions, they use
steam-powered turbines fueled by boil-off natural gas. Movement has begun
toward use of duel-fuel diesel engines, with efficiencies of 38 to 40%, compared
with steam-powered turbines, with efficiencies of 28%. Diesel engines also have
lower NO
X
emissions (Anonymous, 2003).
* Draft is the depth of water necessary to float a ship, and freeboard is that part of the ship’s side
between the waterline and the main deck.
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 303
13.5.3.1 Independent Tanks
Independent tanks are self-supporting, do not form part of the ship’s hull, and
are not essential to the hull strength (Marshall, 2002). The principal system in
use today is the Moss system, originally designed by Moss Rosenberg Verft
(see discussions in Marshall [2002] and Harper [2002]). This system uses
placed in the ship’s hull and a view of the vessel with three of the four tanks
installed. Kato et al. (1995) describe a typical design for vessels and tanks used
on two ships for the Indonesia−Japan route. The four aluminum-alloy tanks on
each vessel are of the Moss-4 spherical design and can hold a total of 125,000
m
3
(0.8 MMBbl) of LNG, filled to 98.5% of capacity. The tanks are insulated
on the external surface by an insulation composed of phenol resin and poly-
urethane foam and are designed for a boil-off rate 0.15% per day.
13.5.3.2 Membrane Tanks
Membrane tanks are non self-supporting and consist of a thin metal membrane,
stainless steel or Invar (35% nickel steel), supported by the ship’s hull through
the thermal insulation. Presently, the Invar membrane (Gaz Transport) is more
FIGURE 13.26 Spherical LNG storage tank before installation on carrier hull.
spherical aluminum tanks. Figures 13.26 and 13.27 show a tank ready to be
© 2006 by Taylor and Francis Group, LLC
304 Fundamentals of Natural Gas Processing
popular than the stainless steel (Technigaz) membrane; 24 of the 40 membrane
ships on order will use the Gaz Transport design (Harper, 2002). Invar owes its
popularity to the fact that it has a very small coefficient of thermal expansion in
the operating ranges of the tanks, which are approximately −260 to +180°F
FIGURE 13.27 Ship with three of four storage tanks installed.
FIGURE 13.28 Gaz transport and Technigaz containment systems (adapted from Harper,
2002).
Plywood boxes filled with
perlite insulation
0.7 mm thick invar
(36% nickel steel)
Ship's hull
Gaz transport containment system
Technigaz MK III containment system
Polyurethane foam
1.2 mm stainless steel
Triplex (sandwich of thin
aluminum foil and epoxy)
Polyurethane foam
Ship's hull
Plywood boxes filled with
perlite insulation
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 305
the Technigaz containment systems (Harper, 2002).
13.6 REGASIFICATION AND COLD UTILIZATION
OF LNG
13.6.1 REGASIFICATION
Regasification of the stored LNG is the final step in the operation of LNG peak
shaving and off-load terminal storage facilities. The regasification or vaporization
is accomplished by addition of heat from ambient air, ambient water, or integral-
fired or remote-fired vaporizers. The cost of the regasification system generally
represents only a small fraction of the cost of the storage plant; however, reliability
of the system is most important because failure or breakdown would defeat the
purpose of the facility. Assuming LNG to be pure methane, the energy required to
gasify the liquid is almost 40% of the gross heating value.
The regasification section of a peak shaving plant is designed for only a few
days of operation during the year to meet the extreme peak loads. To obtain adequate
reliability, total send-out capacity may be divided into several independent parallel
systems, each capable of handling all or a substantial fraction of the total demand.
Designs for the regasification section at the offload terminal are similar to those for
a peak shaving plant, but the capacities of these facilities are much higher, and many
more spare items of equipment may be required to achieve adequate reliability.
system. Liquid is pumped from the storage container to the vaporizer. The
pump discharge pressure must be high enough to provide the desired gas
pressure for entry into the transmission or distribution system. Heat is added
to vaporize the high-pressure LNG and to superheat the gas. Gas leaving the
vaporizer must be odorized because the liquefaction process removes any
odorant originally in the gas.
13.6.2 COLD UTILIZATION
A striking characteristic of the LNG industry is the waste of the vast amount of
potential refrigeration available (as noted earlier, about 40% of the gross heating
value). Accordingly, the problem of utilization of the refrigeration from LNG
FIGURE 13.29 Simplified flow diagram of a typical LNG regasification system.
Natural gas
to pipeline
LNG
Liquid pump
Vaporizer
Heat
Odorizer
(−162 to +82°C). Figure 13.28 shows some details of both the Gaz Transport and
Figure 13.29 is a simplified flow diagram of a typical LNG regasification
© 2006 by Taylor and Francis Group, LLC
306 Fundamentals of Natural Gas Processing
vaporization has been examined for years, but making economically viable use
of the resource has been difficult. Two general approaches have been advanced
for utilization of the cold in LNG vaporization:
• Extraction of work from a power cycle that uses the LNG cold as a
heat sink
• Use of the LNG as a source of refrigeration
Only four of the more than 100 large LNG facilities in the world use the cold
in LNG, and none are in the United States. These four commercial facilities are:
• An LNG receiving facility in Barcelona, Spain, which uses an LNG
cryogenic process to recover heavy hydrocarbons from Libyan LNG
(Rigola, 1970)
• An LNG receiving facility at La Spezia, Italy (Picutti, 1970)
• A Japanese facility that produces liquid oxygen and liquid nitrogen
and provides warehouse refrigeration for frozen food as well (Kataoka
et al., 1970); the plant received a 38% reduction in power requirement
by use of the cold LNG
• An LNG-assisted air separation plant in France (Grenier, 1971)
With the expected increase in LNG production over the next decade, there
will be more incentive to utilize the refrigeration available.
13.7 ECONOMICS
A report by the Energy Information Agency (2003b) provides the most recent
information on global LNG costs and, unless otherwise noted, is the source of
all data in this section. The four steps for bringing gas to the consumer are:
• Gas production. Transfer of the gas from the reservoir to the LNG plant
(including any necessary processing) is 15 to 20% of the total cost.
• LNG plant. Treating, liquefaction, NGL recovery, storage, and LNG
loading are 30 to 45% of the total cost.
• LNG shipping. Maritime shipping is 10 to 30% of the total cost.
• Receiving terminal. Unloading, storage, regasification, and distribution
are 15 to 25% of the total cost.
We briefly look at some of the component costs of the last three items.
13.7.1 LIQUEFACTION COSTS
Liquefaction is the largest cost component of the LNG cost train and is typically
high relative to other energy projects, for reasons such as:
• Remote locations
• Strict design and safety standards
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 307
• Large amount of cryogenic material required
• Tendency to overdesign to ensure supply security
An estimate of generic liquefaction costs is $1.09 /MMBtu ($1.03/TJ) (based
on fuel value) for a two-train, 8 million tpy* (tonne-per-year) Greenfield** project
and $0.97 /MMBtu ($0.92/TJ) for an expansion train.
Major savings have been realized over the years by increasing the size of
liquefaction trains, which requires fewer trains for a given output. In the early
days of the industry, trains were 1.0 to 2.0 million tpy, but presently a 7.8-million
tpy train is planned for Qatar.
13.7.2 SHIPPING COSTS
Costs are determined by the daily charter rate, which can vary widely from as
low as $27,000 to as high as $150,000. The average rate for long-term charters
is between $55,000 and $65,000 per day. Table 13.7 lists some representative
shipping rates.
Prices are based on a 138,000 m
3
(0.87 MMBbl)

carrier at a charter rate of
$65,000 per day. Purchase prices for LNG carriers vary significantly, but for a
138,000 m
3
carrier in the mid -1980s, the price was approximately $280 million
(nominal), whereas in November 2003, it had dropped to approximately $155
million. The main factor driving down the price is the increase in the number
of LNG ships over the past 12 years and clearly documents the dramatic
decrease in cost.
TABLE 13.7
Representative LNG Shipping Rates in $/MMBtu ($/TJ)
to LNG terminals the United States
Exporter Everett Cove Point Elba Island Lake Charles
Algeria 0.52 (0.49) 0.57 (0.54) 0.60 (0.56) 0.72 (0.68)
Nigeria 0.80 (0.75) 0.83 (0.78) 0.84 (0.79) 0.93 (0.88)
Norway 0.56 (0.53) 0.61 (0.57) 0.64 (0.60) 0.77 (0.72)
Venezuela 0.34 (0.32) 0.33 (0.31) 0.30 (0.28) 0.35 (0.33)
Trinadad and Tobago 0.35 (0.33) 0.35 (0.33) 0.32 (0.30) 0.38 (0.36)
Qatar 1.37 (1.29) 1.43 (1.35) 1.46 (1.38) 1.58 (1.49)
Australia 1.76 (1.66) 1.82 (1.72) 1.84 (1.74) 1.84 (1.74)
Prices are based on a 138,000 cubic-meter carrier at a charter rate of $65,000 per day
* The term tpy denotes metric ton per year. A ton is equivalent to 48,700 scf (1,380 Sm
3
).
** A Greenfield project is one built in an area without supporting infrastructure.
of shipyards that build the vessels. Figure 13.30 shows the construction price
© 2006 by Taylor and Francis Group, LLC
308 Fundamentals of Natural Gas Processing
13.7.3 REGASIFICATION TERMINAL COSTS
Terminal costs vary widely and naturally are site specific. A small terminal might
cost as little as $100 million, whereas a state-of-the-art Japanese facility can be
$2 billion or more. In the United States, prices are in the range of $300 million
to $400 million for capacities of 183 to 365 Bcf (5.2 to 10.3 Bm
3
) (McCartney,
2005). The price of regasification is generally assumed to add about $0.30/
MMBtu ($0.28/TJ) to the cost of imported LNG.
13.8 PLANT EFFICIENCY
Once a liquefaction plant is operational, a key economic number is plant thermal
efficiency, h, defined as
Table 13.8 summarizes thermal efficiencies for several large baseload plants. Note
that the thermal efficiencies of Table 13.8 are not the thermodynamic efficiencies.
FIGURE 13.30 Construction price of LNG carriers. Price reflects a 125,000 m
3
(0.78 MMBbls)
carrier from 1992 to 2000 and 138,000 m
3
(0.87 MMBbls) carrier from 2001 to 2003 (Energy
Information Administration, 2003b).
TABLE 13.8
Thermal Efficiencies of Baseload Plants
Plant Efficiency (%) Reference
Badak (Indonesia) 87.6 Mahfud and Sutopo (1989)
Bintulu (Malaysia) 90.9 Peres and Punt (1989)
North West Shelf (Australia) 90.0 Bogani (1992)
C
o
s
t

p
e
r

c
a
r
r
i
e
r
,

M
M
$


U
.
S
.

(
n
o
m
i
n
a
l
)
0
50
100
150
200
250
300
1
9
9
2
1
9
9
3
1
9
9
4
1
9
9
5
1
9
9
6
1
9
9
7
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
η
+ (LNG delivered associated condensate) in Btu
(natural gas delivered) in Btu

]
]
]
]]
×100
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 309
13.9 SAFETY AND ENVIRONMENTAL
CONSIDERATIONS
The major difference in safety of LNG facilities compared with that of normal
gas plants is the larger equivalent amount of gas present on ships and in storage
facilities. As noted earlier, current large LNG carriers hold the equivalent of 4.3 Bcf
(120 MMSm
3
). We first discuss two aspects relating to ships, and then touch on
the controversy that surrounds the proposed new LNG terminals.
An aspect of LNG marine transport that has received much attention is safety.
In addition to the usual problems in handling large quantities of flammable
material, two safety problems are unique to LNG, and both are a function of the
low temperature of the product.
The first problem involves materials of construction. Material for tanks and
cargo-handling equipment must be chosen with care because of thermal expan-
sion and contraction problems during the normal thermal cycling of the equip-
ment. Many materials, notably carbon steel, become dangerously brittle at low
temperatures and are unsatisfactory for normal use. This problem has been
resolved as newer and bigger LNG carriers have been brought into service.
According to the Federal Energy Regulatory Commission (FERC 2005) more
than 33,000 LNG carrier voyages have taken place in the past 40 years, with
no significant accidents.
The second problem is the so-called LNG−water pressure shock wave
that may occur when large quantities of LNG are rapidly dumped on the
ocean, as would be the case in a maritime accident. This phenomenon received
much attention (Burgess et al., 1970; Nakanishi and Reid, 1971) because of
its unique character. Nakanishi and Reid (1971) give an excellent summary
of the problem and suggest an explanation based on the rapid, often violent
boiling caused by the extreme large temperature difference between the LNG
and the water. In more recent work, Cleaver et al. (1998) summarize the
results of a collaborative industrial research program on the rapid phase
transitions of LNG, and Conrado and Vesovic (1998) present the results of a
study on the effect of composition on the vaporization rate of LNG on water.
Proposals have been made to establish at least 55 new LNG receiving
facilities in North America (FERC, 2005). The most desirable locations are in
areas of high population density (e.g., the northeast), so that pipeline costs are
minimized. However, the combination of the perceived threat from terrorists
and of officials not wanting any new facilities in local areas has slowed the
implementation of new receiving facilities. An alternative is to locate the receiv-
ing facilities offshore. A project is underway to convert an existing LNG carrier
into a floating storage and regasification unit (FRSU) (Anonymous, 2006). The
facility, scheduled for completion in 2007, will be capable of delivering 2.75
BSm
3
/yr ashore via subsea pipeline. Existing facilities plan to increase storage
capacity from 18.8 to 35.2 Bcf (532 to 997 MMSm
3
). Most of the expansion
is on the Gulf Coast, far from the major population centers. This expansion
might meet estimated demands up to 2008.
© 2006 by Taylor and Francis Group, LLC
310 Fundamentals of Natural Gas Processing
To meet the anticipated demands for energy and for LNG imports, people on
all sides of the issues must work together to determine the true risks and decide
if energy supplies are worth the risks.
REFERENCES
Acton, A., and Van Meerbeke, R.C., Rollover in LNG Storage—An Industry Perspective,
Proceedings of the Eighth International Conference on Liquefied Natural Gas,
Gas Technology Institute, Chicago, IL, 1986.
Ahlin, L. and Lindblom, U., A New Swedish Concept for Storing LNG Underground,
Proceedings of the Tenth International Conference on Liquefied Natural Gas, Gas
Technology Institute, Chicago, IL, 1992.
Andress, D.L., The Phillips Optimized Cascade LNG Process, A Quarter Century of Improve-
Anonymous, LNG Liquefaction cycles, Cryogen. Industrial Gases, 5 (5) 33, 1970.
Anonymous, Future LNG Tanker Design Firming Up? Nav. Architect, Nov, 3, 2003,
Anonymous, LNG carrier to become a storage, regas unit, Oil Gas J., 104.4, 10, 2006.
Barron, R., Cryogenic Systems, McGraw-Hill, New York, 1966.
Blakely, R., Remote areas of Canada can now be served by trucked LNG, Oil Gas J., 66
(1) 60,1968.
Bogani, F., Initial Experience with the NWS LNG Plant, Proceedings of the Tenth International
Conference on Liquefied Natural Gas, Gas Technology Institute, Chicago, IL, 1992.
Boulanger, A. and W. Luyten, W., Test Cryogenic Storage at Schelle, Results and Con-
clusions, Proceedings of the Seventh International Conference on Liquefied
Natural Gas, Gas Technology Institute, Chicago, IL, 1983.
Burgess, D.S., Murphy, J.N., and Zebetakis, M.G, Hazards Associated with the Spillage
of Liquefied Natural Gas on Water, U.S. Bureau of Mines Report of Investigations
7448, 1970.
California Energy Commission, Algerian LNG Plant Explosion Fact Sheet, 2004,
August 2005.
Cleaver, P., C., et al., Rapid Phase Transition of LNG, Proceedings of the Twelfth International
Conference on Liquefied Natural Gas, Gas Technology Institute, Chicago, IL, 1998.
Closner, J.J., Prestressed Concrete Storage Tanks for LNG Distribution, Conference Pro-
ceedings, American Gas Association, May 8, 1968.
Collins, S. C., A helium cryostat, Rev. Sci. Instr., 18, 157, 1947.
Conrado, C., and Vesovic, V., Vaporization of LNG on unconfined water surfaces, Pro-
ceedings of the Twelfth International Conference on Liquified Natural Gas, Gas
Technology Institute, Chicago, IL, 1998.
Dimentberg, M., Development of LNG pipeline technology, Cryogen. Industrial Gases,
6 (5) 29, 1971.
Dodge, B. F., Chemical Engineering Thermodynamics, McGraw-Hill, New York, 1944.
Dragatakis, L., and Bomard, H., Revythousa Island LNG facility-LNG Storage Tanks
for Seismically Highly Affected Areas, in Proceedings of the Tenth International
Conference on Liquefied Natural Gas, Gas Technology Institute, Chicago, IL,
1992.
ments, 1996, http://lnglicensing.conocophillips.com/NR/rdonlyres/FBB 538DA-
256D-4B96-A844-5D147F4441CF/0/quartercentury.pdf, Retrieved October 2005.
www.rina.org.uk/rfiles/navalarchitect/nov03editorial.pdf, Retrieved August 2005.
www.energy.ca.gov/lng/news_items/2004-01_algeria_factsheet.html, Retrieved
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 311
Duffy, A.R., et al., Heat transfer characteristics of belowground storage, Chem. Eng.
Progress, 63, 55, 1967.
Energy Information Agency, U.S. Department of Energy, U.S. LNG Markets and Uses,
Natural Gas Market: Status And Outlook, DOE/EIA-0637, 2003b,
ber 2005.
Energy Information Administration, U.S. Department of Energy, U.S. LNG Markets and
Engineering Data Book, 12th ed., Sec. 2, Product Specifications, Gas Processors Supply
Association, Tulsa, OK, 2004a.
Engineering Data Book, 12th ed., Sec. 13, Compressors and Expanders, Gas Processors
Supply Association, Tulsa, OK, 2004b.
Engineering Data Book, 12th ed., Sec. 14, Refrigeration, Gas Processors Supply Associ-
ation, Tulsa, OK, 2004c.
Ffooks, R.C. and Montagu, H.E., LNG ocean transportation: experience and prospects,
Cryogenics, 7, 324, 1967.
Finn, A.J., Johnson, G.L., and Tomlinson, T.R., LNG Technology for Offshore and Mid-
Scale Plants, Proceedings of the Seventy-Ninth Annual Convention of the Gas
Processors Association, Tulsa, OK, 2000.
Foglietta, J.H., Production of LNG Using Dual Independent Expander Refrigeration
Cycles, Proceedings of the Eighty-First Annual Convention of the Gas Processors
Association, Tulsa, OK, 2002.
Foerg, W., et al., A New LNG Baseload Process and the Manufacturing of the Main Heat
Exchangers, Proceedings of the Twelfth International Conference on Liquefied
Natural Gas, Institute of Gas Technology, 1998.
Forg, W. and Etzbach, V,. Linde Rep. Sci. Technol., 15, 27, 1970.
General Accounting Office, Report to the Congress of the United States, Liquefied Energy
Giribone, R., and J. Claude, J., Comparative Safety Assessment of Large LNG Storage
Tanks, Chapter 10 Proceedings of the Eleventh International Conference on Liq-
uefied Natural Gas, Gas Technology Institute, Chicago, IL, 1995.
Goodwin, R.D., Thermophysical Properties of Methane, from 90 to 300 K at Pressures to
700 Bar, NBS Technical Note 1653, April, 1974.
Grenier, M.R., La Centrale d’Oxygene de Fos-sur-Mer, Proceedings of the XIII Congress
of the International Institute of Refrigeration, Washington, D.C., 1971.
Gudmundsson, J.S. and Mork, M., Stranded gas to hydrate for storage and transport, 2001
2005.
Hale, D., San Diego LNG plant puts “pressure energy” to work, Am. Gas J. 193 (1) 30,
1966.
International Gas Research Conference, Amsterdam, November, 2001,http://www.
ipt.ntnu.no/~ngh/library/paper11/Amsterdam2001Stranded.htm, Retrieved September
2003a, www.eia.doe.gov/pub/oil_gas/natural_gas/feature_ articles/20 03/lng/lng
Energy Information Administration, U.S. Department of Energy, The Global Liquefied
2003.pdf, Retrieved October 2005.
www.eia.doe. gov/oiaf/analysispaper/global/pdf/eia_0637.pdf, Retrieved Octo-
Uses: June 2004 Update, 2004, www.eia.doe.gov/pub/oil_gas/natural_gas/feature
_articles/2004/lng/lng2004.pdf, Retrieved June 2005.
Federal Energy Regulatory Commission, U.S. Department of Energy, www.ferc.gov/
industries/lng.asp, Retrieved October 2005.
Groupe International des Importateurs de Gaz Natural Liquefie (G.I.I.G.N.L), www.giignl .org,
Gases Safety, Chapter 10, 1978.
undated, Retrieved August 2005.
© 2006 by Taylor and Francis Group, LLC
312 Fundamentals of Natural Gas Processing
Harper, I., Future Development Options for LNG Marine Transportation, Second Topical
Conference on Natural Gas Utilization, Conference Proceedings, AIChE Spring
National Meeting, March 10 – 14, 2002.
Hathaway, P. and Lofredo, A., Experience pays off for this peak shaving pioneer, Cryogen.
Industrial Gases, 6 (5) 25, 1971.
Hidayati, L., Marlono, Y., and Mattteighianti, M., Comparison of LNG Transportation to
JAWA Island: Tanker or Gas Pipeline, Proceedings of the Twelfth International
Conference on Liquefied Natural Gas, Gas Technology Institute, Chicago, IL, 1998.
Hoover, T.E., Technical Feasibility and Cost of LNG Pipelines, Presented at the Second
International Conference and Exhibition on LNG, Paris, 1970.
Jackson, G. and Powell, J., A Novel Concept for Offshore LNG Storage Based on Primary
Containment in Concrete, Proceedings of the Thirteenth International Conference
on Liquefied Natural Gas, Gas Technology Institute, Chicago, IL, 2001.
Kataoka, H., Fujisawa, S., and Inoue, A., Utilization of LNG Cold for the Production of
Liquid Oxygen and Liquid Nitrogen, Proceedings of the Second International
Conference and Exhibition on LNG, Paris, 1970.
Kato, M., et al., Flora and Vesta LNG Carrier Construction, Proceedings of the Eleventh
International Conference on Liquefied Natural Gas, Gas Technology Institute,
Chicago, IL, 1995.
Katz, D.L. and H.T. Hashemi, Pipeline LNG in the Dense Phase, Oil Gas J., 69 (23) 55,1971.
Kotzot, H., Gas Treatment, Liquefaction and Storage: LNG Seminar, Proceedings of the
Eighty-Second Annual Convention of Gas Processors Association, Tulsa, OK,
2003.
Legatos, N.A., et al., Very Large Prestressed Concrete LNG Tanks, Proceedings of the
Eleventh International Conference on Liquefied Natural Gas, Gas Technology
Institute, Chicago, IL, 1995.
Leibon, I., Davenport, S.T., and Muenzler, M.H., Comparison of Natural Gas Transpor-
tation Using LNG Pipelines, and Other Methods, Proceedings of the Eighth
International Conference on Liquified Natural Gas, Gas Technology Institute,
Chicago, IL, 1986.
Linnett, D. T. and Smith, K.C., Mixed refrigerant cascade liquefiers for natural gas, design
and optimization, Advances in Cryogenic Engineering, Vol. 19, Timmerhaus, K.D.
Ed., Plenum Press, New York, 1970, p. 18.
LNG Observer, Oil Gas J., 103.37, 37, 2005.
Long, B., Bigger and Cheaper LNG Tanks? Overcoming the Obstacles Confronting Free-
standing 9% Nickel Steel Tanks up to and Beyond 200,000 m
3
, Proceedings of
the Twelfth International Conference on Liquefied Natural Gas, Gas Technology
Institute, Chicago, IL, 1998.
Mahfud, H. and Sutopo, Availability and Efficiency Improvement of Badak LNG Plant,
Proceedings of the Ninth International Conference on Liquefied Natural Gas, Gas
Technology Institute, Chicago, IL, 1989.
Marshall, G., LNG Ships—An Important Link in the Chain, Second Topical Conference
on Natural Gas Utilization, Conference Proceedings, AIChE Spring National
Meeting, March 10−14, 2002.
McCartney, D.G., Gas Conditioning For Imported LNG, Proceedings of the Eighty-Second
Annual Convention of the Gas Processors Association, Tulsa, OK, 2003.
McCartney, D.G., private communication, 2005.
Miller, R.W. and Clark, J.A., Liquefying natural gas for peak load supply, Chem. Metal-
lurgical Eng. 48 (1) 74, 1941.
© 2006 by Taylor and Francis Group, LLC
Liquefied Natural Gas 313
Monfore, G.E. and Lentz, A.E., Physical Properties of Concrete at Very Low Temperatures,
Portland Cement Association Research Bulletin No.145, Chicago, IL, 1962.
NFPA 59A. Standard for the Production, Storage, and Handling of Liquefied Natural Gas
(LNG) (2006), National Fire Protection Association, 2005.
Nakanishi, E. and Reid, R.C., Liquid natural gas-water reaction, Chem. Eng. Progress., 67 (12)
36, 1971.
Nishizaki, T., et al., Largest Aboveground PC LNG Storage Tank in the World, Incor-
porating the Latest Technology, Proceedings of the Thirteenth International
Conference on Liquefied Natural Gas, Gas Technology Institute, Chicago, IL,
2001.
Okamoto, H. and Tamura, K., Hatsukaichi LNG Receiving Terminal Construction-LNG
In-Pit Storage Tank Construction, Proceedings of the Eleventh International Con-
ference on Liquefied Natural Gas, Gas Technology Institute, Chicago, IL, 1995.
Peres, R. and Punt, A.R., Step by Step Optimization Of Bintulu LNG Plant, Proceedings
of the Ninth International Conference on Liquefied Natural Gas, Gas Technology
Institute, Chicago, IL, 1989.
Perez, V., et al., The 4.5 MMTPA LNG train—A Cost Effective Design, Proceedings of
the Twelfth International Conference on Liquefied Natural Gas, Gas Technology
Institute, Chicago, IL, 1998.
Picutti, E., The LNG Terminal at La Spezia, Proceedings of the Second International
Conference and Exhibition on LNG, Paris, 1970.
Prater, P. B., Design considerations of an LNG peak shaving facility, Cryogen.Industrial
Gases, 5 (3) 15, 1970.
Price, B.C., Winkler, R., and Hoffart, S., Developments in the Design of Compact LNG
Facilities, Proceedings of the Seventy-Ninth Annual Convention of the Gas Pro-
cessors Association, Tulsa, OK, 2000, 407.
Rapallini, R. and Bertha, G.J., Conversion of the LNG Storage Tanks at Panigaglia, Italy,
to Double Containment Storage Systems, Proceedings of the Twelfth International
Conference on Liquefied Natural Gas, Gas Technology Institute, Chicago, IL,
1998.
Rigola, M. Recovery of Cold in a Heavy LNG, Proceedings of the Second International
Conference and Exhibition on LNG, Paris, 1970.
Romanow, S., Got gas? Hydrocarbon Process., 80 (11) 11, 2001.
Salama, C. and Eyre, D.V., Multiple refrigerants in natural gas liquefaction, Chem. Eng.
Progr., 63 (6) 62, 1967.
Smith, J.M., Van Ness, H.C., and Abbott, M.M., Introduction to Chemical Engineering
Thermodynamics, 6th ed.,, McGraw-Hill, New York, 2001.
Swearingen, J. S., Engineers Guide to Turboexpanders, Hydrocarbon Processing, April,
97, 1970.
Taylor, M., Alderson, T., and Mounsey, P., LNG Is King—But What Are the Alternatives?
Proceedings of the Thirteenth International Conference on Liquefied Natural Gas,
Gas Technology Institute, Chicago, IL, 2001.
Troner, A., Russian Far East natural gas searches for a home, Oil Gas J., 99.10, 68, 2001.
Umemura, J., et al., Technological Challenges for the Construction of the Ohgishima LNG
Terminal, Proceedings of the Twelfth International Conference on Liquefied
Natural Gas, Gas Technology Institute, Chicago, IL, 1998
Vink, K.J. and Nagelvoort, R. K., Comparison of Baseload Liquefaction Processes, Pro-
ceedings of the Twelfth International Conference on Liquefied Natural Gas, Gas
Technology Institute, Chicago, IL, 1998.
© 2006 by Taylor and Francis Group, LLC
314 Fundamentals of Natural Gas Processing
Wagner, J.V., Marine Transportation Of Compressed Natural Gas, Proceedings of the
Second Topical Conference on Natural Gas Utilization, AIChE National Meeting,
Spring 2002.
Wagner, J.V. and Cone, R.S., Floating LNG Concepts, Proceedings of the Eighty-Third
Annual Convention of the Gas Processors Association, Tulsa, OK, 2004.
Williams, V.C., Cryogenics, Chem. Eng. 77, 92, 1970.
WEB SITES
mation on current issues relating to LNG. It has many links to other related sites.
Operated by the American Petroleum Institute.
at the state energy commission. Provides an overview of LNG beyond California
and contains many links.
to LNG on worldwide basis.
The Center for LNG, www.lngfacts.org/: Homesite for organization that provides infor-
California Energy Commission, www.energy.ca.gov/lng/index.html: Home page for LNG
Federal Energy Regulatory Commision, U.S. Department of Energy, www.ferc.gov/
LNG Journal home page, www.lngjournal.com/: Page contains recent news stories related
industries/lng.asp.
© 2006 by Taylor and Francis Group, LLC
315
14
Capital Costs of Gas
Processing Facilities
14.1 INTRODUCTION
Completely objective capital cost data for any kind of major plant facilities, for
obvious reasons, are rarely available in the open literature. However, two papers
by Tannehill (2000, 2003) provide estimated capital costs for grass-root facilities.
These data are unique in that they represent consensus values based upon input
from numerous engineering firms or actual plant cost data. With the author’s
permission, the data are presented here.
Numerous assumptions and qualifications to the costs are outlined below.
These cost data should be used only as a quick guide to a first estimate of capital
costs. An added contingency of 20 to 40% is realistic.
After a discussion on the basic premises that apply to all cost data, this chapter
presents cost data along with the premises for the costs of the major components
in a gas plant. The chapter concludes with some complete plant costs.
14.2 BASIC PREMISES FOR ALL PLANT COMPONENT
COST DATA
in this chapter for plant components. These data apply only to new, grass-root
facilities, and are not applicable to retrofits or used equipment. Tannehill (2000)
estimates that costs excluded in Table 14.1, plus other contingencies, constitute
an additional 25 to 40% to the total costs. All costs are based upon U.S. dollars
in either 1999 or 2002.
14.3 AMINE TREATING
of acid gases (CO
2
and H
2
S) as function of plant capacity. The costs assume the
following:

• Operating pressure of 600 to 800 psig (41 to 55 barg)
Table 14.1 presents the basic premises that apply to all of the cost data presented
Figure 14.1 shows the capital cost for new facilities that remove various amounts
Use of DEA (see Chapter 5) as solvent
© 2006 by Taylor and Francis Group, LLC
316 Fundamentals of Natural Gas Processing
The costs exclude any dehydration of gas after treating. Also, no costs are
allocated for BTEX containment. Operating and capital costs increase with acid gas
volume to be removed because both absorber and regenerator capacities must
increase to handle increased amine recirculation rates.
TABLE 14.1
Premises and Assumptions for All Capital Cost Data
Costs Include Costs Exclude
U.S. Gulf Coast location Miscellaneous equipment associated with
grass-roots plant
Those directly associated with process Costs not directly associated with process
Two-month startup operating expense Site and site preparation
Initial supplies and minimum spare parts Owner home-office costs
Sales taxes Interest on investment during construction
Contingency of 10% Construction and bond costs
New facilities only
Source: Tannehill (2000).
FIGURE 14.1 Capital cost of gas treating by use of DEA as a function of plant capacity
in 1999. The lines denoted by x, ,, denote 2, 5, 10, and 20% acid gas removal,
respectively. See Table 14.1 and text for premises (Tannehill, 2000).
0.1
1.0
10.0
100.0
0 200 400 600 800 1000 1200
Plant capacity, MMscfd
C
a
p
i
t
a
l

c
o
s
t
,

M
M
$
© 2006 by Taylor and Francis Group, LLC
Capital Costs of Gas Processing Facilities 317
14.4 GLYCOL DEHYDRATION
These costs are based upon the following:
• Use of TEG
• Having BTEX containment equipment
• Operating pressure of 600 to 800 psig (41 to 55 barg)
The cost is for a standard glycol unit and does not include enhancements to
yield lean glycol concentrations of greater than about 98.6 wt%. Adding the
capital costs for both amine and glycol processes gives the overall cost for gas
treating and dehydration.
14.5 NGL RECOVERY WITH STRAIGHT
REFRIGERATION (LOW ETHANE RECOVERY)
produce a single NGL product stream. The figure shows how an increase in the
GPM (on C
3
+ basis) affects the cost. Costs include limited storage and use of
ethylene-glycol injection for hydrate inhibition. Glycol regeneration costs are
included. However, the costs exclude any cost of upstream compression or
upstream treating. Costs also exclude any fractionation of the liquid product but
do include a stripping column to produce a truckable liquid product. It excludes
any outlet-gas compression because the pressure drop is small through the unit.
premises (Tannehill, 2000).
0.1
1.0
10.0
0 100 200 300 400 500 600 700 800 900 1000 1100
Plant capacity, MMscfd
C
a
p
i
t
a
l

c
o
s
t
,

M
M
$
FIGURE 14.2 Capital cost of TEG dehydration in 1999. See Table 14.1 and text for
Figure 14.2 gives the capital cost of new glycol dehydration facilities (see Chapter 6).
Figure 14.3 provides costs for a straight refrigeration process (see Chapter 7) to
© 2006 by Taylor and Francis Group, LLC
318 Fundamentals of Natural Gas Processing
Going from a lean gas (1.5 GPM, 0.2 m
3
/Mm
3
) to a rich gas (6 GPM, 0.8
m
3
/Mm
3
) doubles the capital cost for large plants, primarily because of the
increased refrigeration load required.
14.6 NGL RECOVERY WITH CRYOGENIC
PROCESSING (HIGH ETHANE RECOVERY)
2
based on 1999 costs. The figure also includes revised costs in 2002 dollars.
The figure shows a roughly 10% increase in capital costs between 1999 and
2002. It also shows that including a nitrogen rejection unit (NRU) to a plant with
a capacity of over 100 MMscfd (3 MMSm
3
/d) adds 60% to the cost of the
combined unit.
14.7 SULFUR RECOVERY AND TAIL GAS CLEANUP
14.7.1 HIGH SULFUR RECOVERY RATES
For processing gas that produces more than about 50 long tons per day of elemental
sulfur, modified Claus units, with or without tail gas cleanup, are the most eco-
FIGURE 14.3 Capital cost of NGL recovery with straight refrigeration in 1999. See
GPM on a C
3
+ basis, respectively (Tannehill, 2000).
0.1
1.0
10.0
100.0
0 20 40 60 80 100 120
Plant capacity, MMscfd
C
a
p
i
t
a
l

c
o
s
t
,

M
M
$
Table 14.1 and text for premises. The lines denoted by ,, represent 1.5, 3, and 6
Table 14.2 gives the premises used for the cost data.
Figure 14.4 shows capital costs for recovering C + by use of cryogenic process-
nomical. Figure 14.5 shows capital costs for a Claus sulfur recovery unit, with and
ing, with and without additional cryogenic nitrogen rejection (see Chapter 8),
© 2006 by Taylor and Francis Group, LLC
Capital Costs of Gas Processing Facilities 319
inclusion of a tail gas cleanup unit adds 50 to 90% to the cost of sulfur recovery.
14.7.2 LOW SULFUR RECOVERY RATES
For production of less than about 50 long tons per day of elemental sulfur, a
number of processes that make a direct conversion of H
2
S into elemental sulfur
are economically viable. Amine treating may or may not be used. One commonly
line with symbol includes cryogenic nitrogen recovery, and lower solid line omits
nitrogen recovery. Both are based on 1999 costs (Tannehill, 2000). The dashed line with
open symbols shows revised capital costs in 2002 dollars for recovery without nitrogen
TABLE 14.2
Premises for Cryogenic Processing Capital Costs
Costs Include Costs Exclude
Molecular sieve dehydration Feed gas treating
Limited storage Compression of feed gas
Recompression of residue gas
to plant inlet pressure
Additional fractionation
of demethanizer bottoms product
Recompression or further handling
of nitrogen gas
Source: Tannehill (2000).
1
10
100
1000
0 100 200 300 400 500 600 700 800 900 1000
Plant capacity, MMscfd
C
a
p
i
t
a
l

c
o
s
t
,

M
M
$
FIGURE 14.4 Capital cost of NGL recovery with cryogenic fractionation. Upper solid
without tail gas cleanup (see Chapter 11), as a function of sulfur production.
Table 14.3 gives the premises used for generating the costs. The data show that
rejection (Tannehill, 2002). See Tables 14.1 and 14.2 for premises.
© 2006 by Taylor and Francis Group, LLC
320 Fundamentals of Natural Gas Processing
costs for one process, the autorecirculation Lo-Cat
®
process for sulfur production
rates up to 20 long tons per day. The costs assume the feed is a concentrated H
2
S
FIGURE 14.5 Capital cost of modified Claus sulfur recovery processes, with and without
TABLE 14.3
Premises for Claus Unit Sulfur Recovery Process Capital Costs Along
with Those for Tail Gas Clean up
Costs Include Costs Exclude
For Claus unit
Claus unit with 3 reactor beds and indirect heat Treating natural gas to obtain H
2
S
Feed gas H
2
S content of > 80% with
sulfur recovery rates of 94 to 97%
Process for enriching feed gas
Only boiler and condenser waste heat
Incinerator and limited storage
For tail gas clean up
Amine-based clean up by use of selective solvent
(e.g., MDEA)
Onsite disposal of purge water
Direct-fired heater
Reducing gas generator
Maximum use of air-cooled heat exchangers
Heat recovery by steam generation
Source: Tannehill (2000).
1
10
100
0 50 100 150 200 250 300 350 400 450
Capacity (long tons sulfur per day)
C
a
p
i
t
a
l

c
o
s
t
,

M
M
$
stream coming from a conventional amine treater. Table 14.4 lists the premises
tail gas clean up, in 1999. See Tables 14.1 and 14.3 for premises (Tannehill, 2000).
used technology is liquid redox (see Chapter 11). Figure 14.6 shows the capital
© 2006 by Taylor and Francis Group, LLC
Capital Costs of Gas Processing Facilities 321
for the costs. Incinerator costs could not be included because of the wide variation
in the CO
2
and other inerts in the feed gas.
14.8 NGL EXTRACTION PLANT COSTS FOR
LARGER FACILITIES
Tannehill (2003) provides cost data for grass-root NGL recovery facilities in the
3
The cost premises of Table 14.1 apply. Note the range in costs for the plants at
a given size.
An important factor in all economic studies is estimated time for construction.
It not only is important for planning but also affects economics in terms of when
gives the time from when the construction contract was awarded until startup for
14.4 for premises (Tannehill, 2000).
TABLE 14.4
Premises for Liquid-Redox Sulfur Recovery Process Capital Costs
Costs Include Costs Exclude
Feed gas H
2
S content of >80% Preconditioning or filtering of feed gas
Sulfur melter Treating natural gas to obtain H
2
S
Initial supplies and minimal spare parts Incinerator
Limited storage
Source: Tannehill (2000).
0.1
1.0
10.0
0 10 15 20 25
Capacity ( long tons sulfur per day)
C
a
p
i
t
a
l

c
o
s
t
,

M
M
$
5
FIGURE 14.6 Capital cost of liquid-redox sulfur recovery process. See Tables 14.1 and
75 to 300 MMscfd (2.1 to 8.5 MMSm /d) size range as shown in Figure 14.7.
the plant will produce a revenue, tied up capital, and the interest cost. Figure 14.8
© 2006 by Taylor and Francis Group, LLC
322 Fundamentals of Natural Gas Processing
the plants in the study. The times also show wide variation. Plants smaller than
150 MMscfd (4.2 MMSm
3
/d) took 8 to 14 months to build, whereas the larger
ones took 11 to 18 months. The one plant that took 24 months was a complex
facility and involved large treating and sulfur recovery facilities, in addition to
the NGL recovery facilities (Tannehill, 2003).
FIGURE 14.7 Capital costs of gas plants in the 75 to 300 MMscfd (2.1 to 8.5 MMSm
3
)
FIGURE 14.8 Time of construction for plants in Figure 14.7 (Tannehill, 2003).
0
10
20
30
40
50
60
70
80
0 50 100 150 200 250 300 350
Capacity, MMscfd
C
a
p
e
x
,

M
M
$
'
s
0
6
12
18
24
30
0 50 100 150 200 250 300 350
Capacity, MMscfd
M
o
n
t
h
s

f
r
o
m

a
w
a
r
d

t
o

s
t
a
r
t
-
u
p

range. See Table 14.1 for premises (Tannehill, 2003).
© 2006 by Taylor and Francis Group, LLC
Capital Costs of Gas Processing Facilities 323
14.9 CORRECTIONS TO COST DATA
A number of corrections are required for use of the above cost data for preparation
of preliminary cost estimates. The two major factors are location and inflation.
Providing these data is beyond the scope of this book. However, initial estimates
of inflation can be made by use of cost-index data available in trade journals. These
data normally include only costs for components such as pumps and labor.
Table 14.5 list the approximate breakdown of costs for the various processes
discussed above. Any cost index can be used if values for 1999 and 2002 are known.
REFERENCES
Tannehill, C.C., Budget Estimate Capital Cost Curves for Gas Conditioning and Process-
ing, Proceedings of the Seventy-Ninth Annual Convention of the Gas Processors
Association, Tulsa, OK, 2000, 141.
Tannehill, C.C., Update if Budget Estimate Capital Cost Curves for NGL Extraction with
Cryogenic Expansion, Proceedings of the Eighty-Second Annual Convention of
the Gas Processors Association, Tulsa, OK, 2003.
Tannehill, C.C., private communication, 2005.
TABLE 14.5
Approximate Distribution of Cost Components
for Gas Processing Facilities
Cost Component
% of Total Cost
NGL Recovery
Amine Treating, Dehydration,
Sulfur Recovery
Pumps, compressors, etc. 20% 20%
Electrical machinery 5% 10%
Internal combustion engines 15% 0%
Instruments 5% 5%
Heat exchangers 10% 15%
Misc. equipment (average) 5% 5%
Materials 15% 20%
Labor 25% 25%
Source: Tannehill (2005).
© 2006 by Taylor and Francis Group, LLC
325
15
Natural Gas Processing
Plants
15.1 INTRODUCTION
discusses three modern plants with significantly different feed and product slates:
• A 500 MMscfd (14 MMSm
3
/d) plant with sweet gas feed and 98%
ethane recovery
• A 279 MMscfd (7.9 MMSm
3
/d) plant with sour gas feed and both NGL
and sulfur recovery
• 7.1 MMNm
3
/d (251 MMscfd) plant with sour gas feed, NGL recovery,
and nitrogen rejection
beginning of most chapters. Details come from the open literature. Feed and product
slates change with time, and plant modifications are common. Therefore, current
plant configuration and operating conditions may differ from those discussed here.
15.2 PLANT WITH SWEET GAS FEED AND 98%
ETHANE RECOVERY
Mallet (1988) provides an excellent description of the San Juan gas plant located
capable of processing 500 MMscfd (14 MMSm
3
/d) of low-sulfur gas, with the
emphasis on obtaining extremely high ethane recovery. To accommodate limita-
tions on equipment sizes and to minimize the loss of productivity during equip-
ment downtime, gas is produced in two identical 250 MMscfd (7.1 MMSm
3
/d)
streams. The streams split after the heat exchange section. Like many modern
gas plants, it generates its own power by use of gas turbines and relies heavily
upon centrifugal compressors.
15.2.1 OVERVIEW OF PLANT FEED AND PRODUCT SLATE
The San Juan plant feed is sweet, that is, it contains less than the maximum
for pipeline gas specifications). Table 1 lists the compositions of both the inlet
gas and the products.
Block schematics for each plant follow those shown in Figure 2.1 and at the
in New Mexico. Figure 15.1 shows a block schematic of the plant, which is
The objective of this chapter is to show how the processes discussed in Chapters
3 through 11 are blended together to form a complete processing plant. This chapter
allowable limit of both sulfur compounds and carbon dioxide (refer to Chapter 1
© 2006 by Taylor and Francis Group, LLC
326 Fundamentals of Natural Gas Processing
While recovering 98% of the ethane, the plant also recovers 100% of the C
3
+
for a total liquids production of 42,000 bpd (6700 m
3
/d). The plant is designed
to produce both ethane-propane-butane-condensate (EPBC) liquid and propane-
butane-condensate (PBC) liquid products. Relative gas and liquid prices dictate
the optimum distribution of ethane between the three streams.
15.2.2 COMPRESSION
The plant receives both high-pressure (320 MMscfd [9 MMSm
3
/d], 850 psig [58
barg]) and low-pressure (120 MMscfd [3.4 MMSm
3
/d

], 350 psig [24 barg]) inlet-
gas feed. The 350 psig (24 barg) feed is pressurized to 850 psig (59 barg) by use
of 1,100 hp (0.82 MW) of centrifugal compression and then enters the heat
exchange block with the high-pressure gas.
15.2.3 HEAT EXCHANGE
The combined gas stream is cooled to 80°F (27°C) in a water-cooled heat exchanger
to condense as much water as possible before the stream splits and enters the
two parallel trains. (The gas that enters the plant contains water, but the water is
that enter a gas plant contain water vapor, but standard practice is not to include
water as a component when reporting the inlet feed composition.)
15.2.4 DEHYDRATION
Recovery of more than 60% of the ethane requires cryogenic processing, and,
consequently, water must be removed to a dew point of −150°F (−101°C).
FIGURE 15.1 Block schematic of the San Juan gas plant. The light and heavy lines denote
gas and liquid streams, respectively (Mallet, 1988).
Inlet
compression
Propane
refrigeration
Dehydration
Residue
compression
Deethanizer
Amine
treating
Heat
exchange High pressure
feed
Low pressure
feed
Compression
PBC product
EPBC product
Sales gas
CO
2
Hydrocarbon
recovery
not reported as a component in the inlet feed composition [Table 15.1]. Most gases
© 2006 by Taylor and Francis Group, LLC
Natural Gas Processing Plants 327
Dehydration in this plant is accomplished by use of three molecular sieve adsorption
beds operating in parallel. Two beds are always adsorbing on 16-hour cycles, while
one bed is undergoing regeneration using 14 MMscfd (0.4 MMSm
3
/d) of dehydrated
gas.
15.2.5 PROPANE REFRIGERATION
Because of the relatively high GPM of the feed, the turboexpander is incapable
of supplying all of the necessary refrigeration. Propane refrigeration (propane
chiller) supplies 64% of the plant requirements.
TABLE 15.1
Feed and Product Compositions for San Juan Plant
Gas Compositions (mol%)
Component Feed Gas Sales Gas
N
2
0.55 0.64
CO
2
1.13 0.54
C
1
85.57 98.67
C
2
7.60 0.15
C
3
3.03 0
iC
4
0.53 0
nC
4
0.78 0
C
5
+ 0.81 0
Gallons/Mscf (m
3
[liq]/MSm
3
[gas]) 3.59 (0.481) 0.04 (0.005)
Btu/scf (kJ/MSm
3
) 1,154.9. (43,356) 978.9 (36,749)
Volume, MMscfd ( MMSm
3
/d) 500 (14) 420 (12)
Product Liquids
Component EPBC
a
(mol%) PBC
b
(mol%)
N
2
0 0
CO
2
0 0
C
1
0.9 0
C
2
58.6 0.7
C
3
23.8 54.2
iC
4
4.2 11.2
nC
4
6.1 16.6
C
5
+ 6.4 17.3
a
Liquid mixture of ethane, propane, butane, and condensate.
b
Liquid mixture of propane, butane, and condensate.
Source: Mallet (1988).
© 2006 by Taylor and Francis Group, LLC
328 Fundamentals of Natural Gas Processing
15.2.6 HYDROCARBON RECOVERY
This section of the plant is a proprietary design that uses cold reflux in the
demethanizer. The turboexpander provides the remaining 36% of the required
plant refrigeration. The expander drops the gas pressure from 850 psig (59 barg)
to 350 psig (24 barg) to provide the −150°F (−101°C) feed to the top of the
demethanizer. The turboexpander operates at 14,000 rpm and provides 4,800 hp
(3.6 MW) of residue gas compression.
The demethanizer has 28 valve trays and operates at 350 psig (24 barg) and
temperatures of −150°F (−101°C) at the top to 63°F (17°C) at the bottom. Four
reboiler types, upper side, lower side, bottom, and trim, are used. (A “trim”
reboiler is a second reboiler in series with the bottom reboiler.) By use of cold
reflux in the top of the tower, the overhead residue gas product is 98.7% methane.
The bottoms product is the EPBC stream that feeds the deethanizer and is also
taken as a product stream. Although the CO
2
content of the inlet gas to the plant
is only 1.13%, the CO
2
concentrates in the demethanizer bottoms, and the EPBC
liquid stream contains 5% CO
2
. Consequently, additional purification is required
for the liquid stream.
15.2.7 AMINE TREATING
The EPBC stream from the demethanizer splits, with one branch feeding the
deethanizer and the other flowing to the amine-treating unit. The amine treater
operates at 655 psig (45 barg) and 85°F (29°C) and uses a 30 wt% diethanolamine
solution to remove the CO
2
. Activated alumina beds are used to remove water
and amine in the liquid EPBC before it goes to storage. Rich amine passes through
a hydrocarbon coalescer, flash vessel, and sock filter to remove hydrocarbons and
impurities before being regenerated. This stream must have minimal hydrocar-
bons in it to avoid “stacking” or foaming in the regenerator.
15.2.8 DEETHANIZER
The tower contains 35 valve trays and operates at 400 psig (28 barg), with
overhead and bottoms temperatures of 58°F (14°C) and 220°F (104°C), respec-
tively. The PBC bottoms product goes to storage, whereas the overhead stream
is partially condensed by use of propane refrigeration to provide reflux to the
column. The vapor from a liquid−vapor separator is compressed and joins the
sales gas stream or is sent to the amine treater.
15.2.9 RESIDUE COMPRESSION
The demethanizer overhead is first compressed by the compressor coupled to
the expander. The final sales gas pressure of 850 psig (59 barg) is obtained
by using two 5,000 hp (3.7 MW) gas-driven centrifugal compressors operating
in parallel.
© 2006 by Taylor and Francis Group, LLC
Natural Gas Processing Plants 329
15.3 PLANT WITH SOUR GAS FEED, NGL,
AND SULFUR RECOVERY
The Whitney Canyon plant, located in Wyoming, is an excellent example of a
large plant that processes a sour gas, extracts NGL, and recovers large quantities
of sulfur. Figure 15.2 is a block diagram that shows only the major processes at
Whitney Canyon. The discussion in this section focuses on those processes
emphasized in this book and is consequently very limited. Webber et al. (1984)
present a very detailed discussion of the entire project, and the material presented
here is taken from their paper. This section discusses some of changes in pro-
cessing required when relatively high H
2
S concentrations are present.
15.3.1 OVERVIEW OF PLANT FEED AND PRODUCT SLATE
The plant was dedicated in late 1982 and lies near the producing field. The field
contains 17 wells that produce from four different formations that range in depth
from 9,000 ft (2,700 m) to 16,500 ft (5,000 m). The H
2
S concentration in the gas
varies from 1 to 19%, depending on the well and producing formation. Because of
the large variability of the feed gas, no feed or product stream compositions were
reported. The plant can handle 270 MMscfd (7.6 MMSm
3
/d) of feed and produce
200 MMscfd (5.6 MMSm
3
/d) of sales gas, 13 Mbbl/d (2 Mm
3
/d) of NGL, 6 Mbbl/d
(1 Mm
3
/d) of condensate, and 1,200 long tons /d (1220 tonne/d) of elemental sulfur.
15.3.2 INLET RECEIVING
Both low-pressure and high-pressure gas from the gathering system enter inlet
receiving, where water and condensate are separated from the gas. After removal
FIGURE 15.2 Block schematic of the Whitney Canyon plant (Webber, et al., 1984).
Liquids
processing
Sulfur
recovery
High pressure
gas
High pressure
gas
Low pressure
gas
Low pressure
gas
Alternate gas
sales
Elemental
sulfur
NGL
Sales gas
Water
Hydrocarbon
recovery
Inlet
receiving
Inlet
compression
Gas treating
Dehydration
© 2006 by Taylor and Francis Group, LLC
330 Fundamentals of Natural Gas Processing
of dissolved H
2
S, the water is sent to disposal wells. The condensate and gas
enter the plant in separate lines that operate at approximately 1,000 psig (69 barg)
and 250 psig (17 barg), respectively. An interesting feature of the plant is that all
inlet-receiving facilities are located in a valley 800 feet (240 m) lower in elevation
than the plant to avoid slug flow into the plant.
An important feature of the inlet facility is control of the temperature of gas
and condensate to manage hydrocarbon dew point and prevent hydrate formation.
The paper does not provide details of these processes or the stripping of H
2
S
from the water.
15.3.3 INLET COMPRESSION
The high-pressure gas from inlet receiving goes directly to gas treating, whereas
the low-pressure gas is compressed by two 1,500 hp (1.1 MW) compressors before
it joins the high-pressure stream at gas treating. Another pair of compressors
process gas from condensate stabilization and from the amine treating flash tank,
so that the gas can be blended with the high pressure gas. Both streams are
discussed in following paragraphs.
15.3.4 GAS TREATING
Gas sweetening is accomplished in two identical parallel trains. Each consists
of an amine unit that uses a 35 wt% solution of diethanolamine to remove
the H
2
S to less than one-quarter grain H
2
S/100scf (6 mg/m
3
) and the CO
2
to
a level sufficiently low to prevent freezing in the hydrocarbon recovery pro-
cess. The amine circulation rate for each train is 2,700 gpm (170 l/s), with a
rich-amine loading of 0.75 moles acid gas per mole of amine. This treating
system differs from most in that both high-pressure and low-pressure contac-
tors are used.
In the high-pressure contactor, sour gas contacts the amine at 95°F (35°C) and
1,000 psig (69 barg) and uses both lean and semi-lean amine from the regenerator.
The lean amine enters at the top of the high-pressure contactor. The semi-lean amine
is a partially regenerated stream from a side draw on the regenerator. It enters the
contactor at a point lower than the lean amine. The high-pressure rich amine is
expanded through a power-recovery turbine to 250 psig (17 barg) and goes to a
low-pressure flash tank for removal of dissolved hydrocarbons.
However, the gas from the flash tank also contains H
2
S. This gas goes to a
low-pressure contactor (250 psig, 17 barg), where it contacts lean amine. The
sweet gas then returns to inlet compression.
15.3.5 SULFUR RECOVERY
The acid gas from gas treating goes to sulfur recovery, where it is processed by
use of a two-train, straight-through Claus process for primary (95%) sulfur recov-
ery. The waste-heat boiler from the Claus process provides much of the plant
steam. Each train has two Claus reactors and sulfur condensers.
© 2006 by Taylor and Francis Group, LLC
Natural Gas Processing Plants 331
Final recovery of the sulfur (tail gas cleanup) is accomplished by use of the
cold-bed adsorption (CBA) process, with two CBA reactors per train. The overall
recovery of sulfur is greater than 99%. The effluent from the CBA reactors is
sent to the incinerator for conversion of all remaining sulfur compounds to SO
2
,
which is then vented into the atmosphere through a stack.
The molten sulfur is trucked 32 miles (51 km), in specially designed trailers,
from the plant to a rail terminal. The sulfur then is shipped to Florida for
manufacture of fertilizer. A rail line was not run to the plant because of concerns
for wildlife.
15.3.6 DEHYDRATION
After removal of acid gases in the gas treating section, the gas goes to dehydration.
A glycol dehydrator and molecular sieve adsorption beds are used in series. The
triethylene glycol dehydrator removes the bulk of the water, but it cannot achieve
the necessary purity (less than 0.1 ppm water) for cryogenic processing, so the
second dehydrator, a four-bed molecular sieve unit, is used for final purification.
Three beds are always in service; two dehydrate while one undergoes regenera-
tion, and the fourth bed acts as a spare. When NGL recovery is not required (no
cryogenic units needed), the molecular sieve beds can be bypassed, and the gas
can go to alternate gas sales.
15.3.7 HYDROCARBON RECOVERY
The plant recovers NGL and up to 88% of the ethane in the gas. The plant design
allows operation in either an ethane recovery or ethane rejection mode, depending
on the current economic situation. Refrigeration is obtained exclusively from gas
expansion (the technique is not mentioned in the paper), and the cold fluids go
to a demethanizer to separate methane.
15.3.8 LIQUIDS PROCESSING
This section of the plant has two separate units: condensate stabilization and NGL
treating. Condensate stabilization consists of taking the liquids from inlet receiv-
ing, water washing them to remove salt, and then processing them in a stabilizer.
The stabilizer removes the H
2
S and light-hydrocarbon components to lower the
vapor pressure of the condensate. A sweet stripping gas, combined with a direct-
fired reboiler, is used to reduce the H
2
S level to less than 10 ppm in the condensate.
The H
2
S stripping gas and light hydrocarbons go to inlet compression and then
to gas treating.
The natural gas liquids are amine treated with a 30 wt% diethanolamine solution
to remove trace contaminants, and then dehydrated, refrigerated, and transferred to
storage. To perform satisfactorily, this amine must have lower impurity levels than
required for gas treating.
© 2006 by Taylor and Francis Group, LLC
332 Fundamentals of Natural Gas Processing
15.4 PLANT WITH SOUR GAS FEED NGL
RECOVERY, AND NITROGEN REJECTION
The Hannibal plant, located in Sfax, Tunisia, processes gas from the offshore Miskar
field. In full operation since 1996, it provides 80% of the gas used in Tunisia. Jones
et al. (1999) and Howard (1998) give complete discussions of the plant. Figure 15.3
is a block diagram of the major processing steps.
15.4.1 OVERVIEW OF PLANT FEED AND PRODUCT SLATE
MMNm
3
/d (265 MMscfd) of gas and produces 4.9 MMNm
3
/d (180 MMscfd) of
sales gas. The sales gas specifications include:
• No H
2
S present (no concentration specification given)
• N
2
not to exceed 6.5 mol%
• H
2
O less than 80 ppmw
• Hydrocarbon dew point less than −5°C (23°F)
Reduction of the N
2
to the specified level requires cryogenic processing in
the nitrogen rejection unit (NRU). Consequently, the CO
2
, H
2
O, and BTEX must
be removed to levels low enough to prevent freezing and plugging in the gas−
gas exchangers.
15.4.2 INLET RECEIVING
The feed from the Miskar field that enters inlet receiving contains both the gas and
the associated wet condensate. A slug catcher separates the liquid from the gas.
The liquid goes to liquids processing, and the gas goes to gas treating. The slug
catcher consists of six 48-inch (120 cm) diameter pipes that are 182 m (600 ft) long.
FIGURE 15.3 Block schematic of the Hannibal gas plant. (Jones; et al., 1999; Howard,
1998)
Gas treating
Dehydration
Inlet receiving
Sulfur
recovery
High pressure
gas
Sales gas
Elemental sulfur
NRU and
cold box
Liquids
processing
Condensate
to storage
The design feed gas composition is shown in Table 15.2. The plant processes 7.1
© 2006 by Taylor and Francis Group, LLC
Natural Gas Processing Plants 333
15.4.3 GAS TREATING
An amine unit that uses 50 wt% activated-MDEA solution removes both the H
2
S
and the CO
2
. The H
2
S is removed to sales gas specifications and the CO
2
to 200
ppmv. Some BTEX is also removed by the amine, however, not enough to prevent
solid deposition in the cryogenic unit. Additional removal is required and is
accomplished in the dehydration section.
15.4.4 SULFUR RECOVERY
The acid gases from the amine unit are treated for sulfur removal by use of the
liquid-redox Lo-Cat
®
process. A Claus unit is not justified because the sulfur
production rate is only approximately 22,000 lb/d (10 tonnes/d).
15.4.5 DEHYDRATION
Water removal is accomplished by use of two dehydrators, a triethylene glycol
unit followed by molecular sieve adsorption beds. Although the glycol dehydrator
reduces the load on the adsorption beds, it was designed for removing the bulk
of the BTEX to prevent freezing in the cryogenic NRU. Thus, they turned what
often is considered an environmental problem into an asset. The glycol regenerator
TABLE 15.2
Design Feed Gas Composition
for the Hannibal Plant.
Component Mol%
Nitrogen 16.903
Carbon dioxide 13.588
Hydrogen sulfide 0.092
Methane 63.901
Ethane 3.349
Propane 0.960
i-Butane 0.258
n-Butane 0.286
i-Pentane 0.142
n-Pentane 0.147
Hexane 0.138
BTEX 0.121
C
7
fraction 0.057
C
8
fraction 0.019
C
9
fraction 0.005
C
10
+ 0.001
Source: Jones et al. (1999).
© 2006 by Taylor and Francis Group, LLC
334 Fundamentals of Natural Gas Processing
operates at around 190°C (375°F). The overhead reflux drum runs at 0.2 barg (3 psig)
and 28°C (82°F), where the water and BTEX are separated, and the BTEX goes
to condensate storage.
15.4.6 NRU AND COLD BOX
The cold box contains a series of heat exchangers to condense the heavier
hydrocarbons and two vapor−liquid separators that operate at different tempera-
tures. After heat exchange, the liquid from the warm separator (−42°C, −44°F)
goes to liquids processing, whereas the gas from this separator is chilled, partially
condensed, and sent to the cold separator (−67°C, −89°F). The liquid from the
cold separator ultimately combines with the liquid from the NRU to become part
of the sales gas. The vapor from the cold separator is chilled, condensed, and
expanded through a J-T valve into the NRU. In the NRU, cryogenic distillation
is used to separate the N
2
from the hydrocarbons. However, the paper provides
no details on the NRU. The N
2
is vented, and the sales-gas stream contains less
than 6.5% N
2
, which, thus, meets specifications.
15.4.7 LIQUIDS PROCESSING
In this plant, liquids processing consists of sending the condensate from inlet
receiving and the liquids from the cold box to a distillation system, where the
lighter components are removed to adjust the vapor pressure of the remaining
liquid to a specific value (condensate stabilization). The stabilized condensate
(Reid vapor pressure of 12 psig [0.8 barg]) is then sent to storage. The paper
provides no details on the fate of the gas from the stripper or on the treating and
disposition of the NGL stream.
REFERENCES
Howard, I., Hannibal’s Experiences, Proceedings of the Laurance Reid Gas Conditioning
Conference, 1998, 194.
Jones, S., Lee, S., Evans, M., and Chen, R., Simultaneous Removal of Water and BTEX
from Feed Gas for a Cryogenic Plant, Proceedings of the Seventy-Eighth Annual
Convention of the Gas Processors Association, 1999, 108.
Mallett, M.W., Conoco/Tenneco Gas Plant Meeting the Challenges, Proceedings of the
Sixty-Seventh Annual Convention of the Gas Processors Association, Tulsa, OK,
1988, 150.
Webber, W.W., Petty, L.E., Ray, B.D., and Story, J.G., Whitney Canyon Project, Proceed-
ings of the Sixty-Third Annual Convention of the Gas Processors Association,
Tulsa, OK, 1984, 105.
© 2006 by Taylor and Francis Group, LLC
335
Notation
A = area, ft
2
(m
2
)
C = constant in equation, units depend upon equation; concentration,
unit is use specific; percent clearance
C
P
= heat capacity at constant pressure, Btu/lb mol-°F (kJ/kg-mol-°C)
C
SS
= correction factor for adsorption of unsaturated gas (see Equation 6.6)
C
T
= correction factor adsorption bed temperature (see Equation 6.7)
C
V
= heat capacity at constant volume, Btu/mol-°F (kJ/kg-mol-°C)
D = diffusivity, ft
2
/h (m
2
/s)
D = outside diameter, ft (m)
d = inside diameter, ft (m)
D
P
= particle or droplet diameter, ft(m)
F = factor in equations, units depend upon equation
g = acceleration due to gravity, 32.2 ft/s
2
(9.81 m/s
2
)
f = fraction, dimensionless; fugacity, psia (bar)
h = enthalpy, Btu/lb (kJ/kg)
h
d
= head, ft-lb
f
/lb
m
(N-m/kg)
H = enthalpy, Btu/lb-mol (J/kg-mol)
= Btu/scf
J = molar flux, lb-mole/ft
2
-h (kg-mole/m
2
-s)
K = y/x, equilibrium constant, dimensionless
L = length, ft (m)
m = number of compressor stages, dimensionless; mass, lb (kg)
= mass flow rate, lb/h (kg/s)
MW = molecular weight or molar mass, lb/lb-mol (g/g-mol)
n = number of moles, lb mols (g mols)
= lb-mol/h (kg-mol/h)
N
Re
= Reynolds number, dimensionless
P = pressure, psia (bara)
P = permeability, lb-mol/ft-psia-h (kg-mol/m-bar-s)
p
i
= partial pressure of component i, psia (bar)
PD = piston displacement, ft
3
/min (m
3
/h)
PR = pressure ratio, also called compression ratio, dimensionless
q = heat, Btu/lb (kJ/kg)
q
L
= heat leak into a system, Btu/lb
Q = volumetric flow rate, ft
3
/min (m
3
/h)
R = universal gas constant, 10.73 psia-ft
3
/°R lb-mol (8.314 Pa-m
3
/
K
s = entropy, Btu/lb-°F (J/kg-K)
H
vi
id
m
n
R = relative ratio of K values, dimensionless (see Chapter 5)
K kmol) (see Appendix B for values in other units)
© 2006 by Taylor and Francis Group, LLC
336 Fundamentals of Natural Gas Processing
S = solubility, lb-mol/psia-ft
3
(kg-mol/bara-m
3
)
SpGr = specific gravity, dimensionless
T = absolute temperature, °R (K)
t = temperature, °F (°C)
V = volume ft
3
(m
3
),
v = volume, ft
3
/lb (m
3
/kg) or ft
3
/lb mol (m
3
/g mol)
V
S
= superficial velocity, ft/min (m/s)
V
T
= terminal velocity ft/s (m/s)
W = water content of gas, lb /MMscf (mg/Sm
3
)
w
S
= Shaft work (positive if work done by system), Btu/lb (kJ/kg)
WB = Wobbe number, Btu/scf (kJ/m
3
)
X = weight fraction, dimensionless
x = mol fraction in liquid phase, dimensionless
y = mol fraction in vapor phase, dimensionless
z = compressibility factor, PV/RT, dimensionless
GREEK LETTERS
a
1−2
= selectivity, and is the ratio of permeabilities, P
1
/P
2
in membranes,
or K values K
1
/K
2
, in distillation, dimensionless
g = ratio of heat capacities, C
P
/C
V
or liquid phase activity coefficient,
dimensionless
h = efficiency, dimensionless
Θ = time, min (s)
k = polytropic constant, dimensionless
l = heat of vaporization, Btu/lb (kJ/kg)
m = viscosity, cP (m-Pa-s); chemical potential, Btu/lb-mol (J/g-mol); Joule-
Thomson coefficient, ºF/atm (ºC/ bar)
r = density, lb/ft
3
(kg/m
3
)
j = fugacity coefficient, dimensionless
SUBSCRIPTS
1 = entering
2 = exiting
A = actual
avg = average value
b = base value
c = critical property
g = gas phase
i = component i, initial value, inhibitor
in = inlet
out = outlet
IS = isentropic
© 2006 by Taylor and Francis Group, LLC
Notation 337
m = mixture
MTZ = mass transfer zone
R = reduced value
L = liquid
P = polytropic
R = relative or reduced condition
rg = regeneration
si = sieve
st = steel
V = volumetric or vapor
W= water
SUPERSCRIPTS
L = liquid
Sat = saturation condition
V = Vapor
© 2006 by Taylor and Francis Group, LLC
339
Appendix A
Glossary of Gas Process
Terminology
Most terms in this glossary were obtained from GPA (1977) and the Engineering
Data Book (2004a). Terms marked with an asterisk were added by the authors.
Absorber capacity: The maximum standard cubic feet per day of natural
gas that can be processed through an absorber at specified absorption oil rate,
temperature, and pressure without exceeding pressure drop or other operating
limitations.
Absorption factor: A factor used in engineering calculations that expresses
the propensity for a constituent in natural gas to be absorbed in a liquid solvent.
This factor is generally found in the literature as A = L/KV, where L and V are the
moles of liquid and vapor, respectively, from a tray or an average value for the
section or total absorber. K is similarly the vapor−liquid equilibrium ratio for
the particular component.
Acid gas: The hydrogen sulfide, carbon dioxide, or both contained in or
extracted from gas or other streams.
Allowable: The maximum rate of production from an oil or gas well or group
of wells that is allowed by a state or governing body. The rate is set by rules that
vary among the different states or governing bodies.
Amine: Any of several compounds such as, but not limited to, monoethano-
lamine (MEA), HOC
2
H
4
NH
2
, employed in treating natural gas. The amines are
generally used in water solutions to remove hydrogen sulfide and carbon dioxide
from gas and liquid streams.
*APC(Advanced process control): A “smart” control system trained to
optimize plant operation. It supervises the basic digital control system (DCS).
Associated gas: Natural gas, commonly known as gas-cap gas, which overlies
and is in contact with crude oil in the reservoir. Where reservoir conditions are
such that the production of associated gas does not substantially affect the recov-
ery of crude oil in the reservoir, such gas may be reclassified as nonassociated
gas by a regulatory agency.
*BACT(Best available control technology): Usually required for emissions
from a new or modified commercial emission source, such as a plant or combus-
tion engine, in a clean area.
© 2006 by Taylor and Francis Group, LLC
340 Fundamentals of Natural Gas Processing
*Baseload plant: A liquefaction facility designed to convert gas from a
stranded reserve to LNG for transit.
Barrel: A unit of liquid volume measurement that, in the petroleum industry,
equals 42 U.S. liquid gallons for petroleum or natural gas products measured at
60°F and at equilibrium vapor pressure. Chemicals may be packaged in barrels
or drums that have capacities of 55 U.S. gallons.
bcf: billion cubic feet
Blanket gas: A gas phase in a vessel above a liquid phase. The purpose may
be to protect the liquid from air contamination, to reduce the hazard of detonation,
or to pressurize the liquid. The source of the gas is external to the vessel.
Blow case: A small tank in which liquids are accumulated and drained by
application of gas or air pressure above the liquid level. Such a vessel is usually
located below a pipeline or other equipment at a location where an outside
power source is not convenient for removing the drained liquids. Sometimes
referred to as a drip.
Blowdown: The act of emptying or depressuring a vessel. This term may
also refer to the discarded material, such as blowdown water from a boiler or
cooling tower.
Bottom hole pressure (temperature): The pressure (temperature) measured
in a well at a depth that is at the midpoint of the thickness of the producing zone.
B-P mix: A liquefied hydrocarbon product composed chiefly of butanes and
propane. If it originates from a refinery, it may also contain butylenes and
propylene. More specifically, it conforms to the GPA specifications for commer-
cial B-P mixes as defined in GPA Publication 2140.
Breathing: The movement of vapor in or out of a storage tank because of
the change of level of the storage liquid, a change in the temperature of the vapor
space above the liquid, or of atmospheric pressure.
bs&w: Bottom sediment and water that collects in the bottom of storage
tanks. It is also called tank bottoms.
*BTEX: benzene, toluene, ethyl benzene, and xylene
Butane, commercial: A liquefied hydrocarbon that consists predominately
of butane, butylenes, or both and which conforms to the GPA specification for
commercial butane defined in GPA Publication 2140.
Butane, field grade: A product that consists predominately of normal butane
and isobutane that is produced at a gas processing plant. Also called “mixed butane.”
Butane, normal: In commercial transactions, a product that meets the GPA
specification for commercial butane and, in addition, contains a minimum of 95
liquid vol% normal butane.
Casinghead gas: The natural gas that is produced from oil wells, along with
the crude oil. Also, it is an obsolete term for oil well gas.
Casinghead gasoline: An obsolete term for natural gasoline.
Charcoal test: A test standardized by the American Gas Association and the
Gas Processors Association to determine the natural gasoline content of a given
© 2006 by Taylor and Francis Group, LLC
Appendix A 341
natural gas. The gasoline is absorbed from the gas on activated charcoal and then
recovered by distillation. The test is described in Testing Code 101-43, a joint
AGA and GPA publication.
Chromatography: A technique for separating a mixture into individual com-
ponents by repeated adsorption and desorption on a confined solid bed. It is used
for analysis of natural gas and NGL.
*CIF: Carriage, insurance, and freight (cost delivered to customer).
Claus process: A process to convert hydrogen sulfide into elemental sulfur
by selective oxidation.
*CBM: Coal bed methane.
Color test: A visual test made against fixed standards to determine the color
of petroleum or other product.
Compression ratio: The ratio of the absolute discharge pressure from a
compressor to the absolute intake pressure. Also applies to one cylinder of a
reciprocating compressor and one or more stages of a rotating compressor.
Compression tests: An AGA-GPA test to determine the natural gasoline
product content of natural gas. Refer to AGA-GPA Testing Code 101-43.
Condensate: The liquid formed by the condensation of a liquid or gas; specifi-
cally, the hydrocarbon liquid separated from natural gas because of changes in tem-
perature and pressure when the gas from the reservoir was delivered to the surface
separators. Such condensate remains liquid at atmospheric temperature and pressure.
Condensate gas reservoir: A hydrocarbon reservoir with natural gas that
will yield condensate.
Condensate well: A gas well producing from a condensate reservoir.
Connate water: Water that settled with the deposition of solid sediments and
that has not existed as surface water at atmospheric pressure. Also, water in a
particular formation that fills a portion of the pore space.
Convergence pressure: The pressure at a given temperature for a hydrocar-
bon system of fixed composition at which the vapor−liquid equilibria values of
the various components in the system become or tend to become unity. The
convergence pressure is used to adjust vapor−liquid values to the particular system
under consideration.
Copper strip test: A test that uses a small strip of pure copper to determine
qualitatively the corrosivity of a product.
Cryogenic plant: A gas processing plant that is capable of producing natural
gas liquid products, including ethane, at very low operating temperatures.
Cycle gas: Gas that is compressed and returned to the gas reservoir to
minimize the decline of reservoir pressure.
Cycling: The process in which effluent gas from a gas reservoir is passed
through a gas processing plant or separation system, and the remaining residue
gas is returned to the reservoir. The word “recycling” has also been used for this
function, but it is not the preferred term.
Cushion gas: Refer to definition of “blanket gas.”
© 2006 by Taylor and Francis Group, LLC
342 Fundamentals of Natural Gas Processing
*DCS (Distributed control system): First-level control system used to
manipulate control valves and other devices to maintain a desired setpoint.
DEA unit: A treating system that uses DEA for reduction of hydrogen sulfide,
carbon dioxide, carbonyl sulfide, and other acid gases from sour process streams.
Debutanizer (Deethanizer): A unit of equipment that separates butane
(ethane), with or without lighter components, as debutanizer (dethanizer) over-
head, from a mixture of hydrocarbons and leaves a bottoms product that is
essentially butane (ethane) free.
Degree-day: A unit of temperature and time that shows the difference between
a 65°F (18.3°C) base and a daily mean temperature when the later is less than 65°F.
This temperature difference is the number of degree-days for a particular day.
Demethanizer: A unit of equipment that separates methane and more volatile
components, as demethanizer overhead, from a mixture of hydrocarbons and
leaves a bottoms product that is essentially methane free.
Depropanizer: A unit of equipment that separates propane, with or without
lighter components, as depropanizer overhead, from a mixture of hydrocarbons
and leaves a bottoms product that is essentially propane free.
Disulfides: Chemical compounds containing an —S—S— linkage. They are
colorless liquids, completely miscible with hydrocarbons and insoluble in water
and sweet to the doctor test. Mercaptans are converted to disulfides in treating
processes that employ oxidation reactions.
Doctor test: A qualitative method to detect hydrogen sulfide and mercaptans
in petroleum distillates. The test distinguishes between “sour” and “sweet” products.
Drip gasoline: Hydrocarbon liquid that separates in a pipeline that transports
gas from the well casing, lease separation, or other facilities and drains into
equipment from which the liquid can be removed.
Ebullition: Boiling, as especially applied to remove heat from engine jacket
water, wherein the water is permitted to boil and the evolved vapors are condensed
in air-fin coolers.
*EPC: Engineering, procurement, and construction.
EP-mix (ethane-propane mix): A product that consists of a mixture of
essentially ethane and propane.
Fast cycle unit (quick or short cycle unit): An adsorption plant that has
adsorption cycles of relatively short duration.
Field processing unit: A unit through which a well stream passes before the
gas reaches a processing plant or sales point. Field-processing units may be
separator systems, LTX units, adsorption units, etc.
Field separator: A vessel in the oil or gas field that separates the gas,
hydrocarbon liquid, and water from each other.
*DDC (Direct digital controllers): See APC.
DEA (Diethanolamine): Refer to the definition of “amine.”
Drip: Refer to definition of “blow case.”
Field condensate: Refer to definition of condensate.
© 2006 by Taylor and Francis Group, LLC
Appendix A 343
Formation gas: Gas initially produced from an underground reservoir.
*FPSO: Floating production, storage, and offloading vessel used in offshore
operations.
Freeze valve: A specially constructed and calibrated valve designed and used
solely to determine the water content in the propane product. See ASTM D 2713.
Full well stream: The total flow stream or effluent from a producing well
that contains all the constituents of the reservoir fluids.
Gas cap: The natural gas zone located above an oil zone in a common reservoir.
Gas drive: A manner of producing crude oil or other liquids from the reservoir
wherein the required energy is provided by gas pressure.
Gas injection: The injection of natural gas into a reservoir to maintain or
increase the reservoir pressure or reduce the rate of decline of the reservoir pressure.
Gas lift: A method of bringing crude oil or water to the surface by injecting
gas into the producing well bore.
Gas-oil ratio (GOR): The ratio of gas to liquid hydrocarbon produced from a
well. This ratio may be expressed as standard cubic feet per barrel of stock tank liquid.
Gasoline: A product that by its composition is suitable for use as a fuel in
internal-combustion engines.
Gasoline plant: A natural gas processing plant. This latter preferred term
helps differentiate from a unit that makes gasoline within an oil refinery.
Gas reservoir: A geological formation that contains a single gaseous phase.
When the gas is produced, the surface equipment may or may not contain con-
densed liquid, depending on the temperature, pressure, and composition of the
single-reservoir phase.
Gas-well gas: The gas produced or separated at surface conditions from the
full well stream produced from a gas reservoir.
Gas-well liquids: The liquid separated at surface conditions from the full
well stream produced from a gas reservoir.
Glycol: A group of compounds used to dehydrate gaseous or liquid hydro-
carbons or to inhibit the formation of hydrates. Commonly used glycols are
ethylene glycol (EG), diethylene glycol (DEG), and triethylene glycol (TEG).
gpm: (1) gpm (gallons per minute); the term used to describe a pumping rate
in gallons per minute for a liquid. (2) gpm; preferably Gal/MCF (gallons per
thousand cubic feet): This term refers to the content in natural gas of components
recoverable or that are recovered as liquid products. The preferred form of the
term prevents confusion with “gallons per minute.”
Gravity, API: An arbitrary scale that expresses the relative density of liquid
petroleum products. The measuring scale is calibrated in degrees API, which is
calculated by the following formula: °API = [141.5/(Sp Gr. 60°F/60°F)] − 131.5.
Gas condensate: Refer to definition of “condensate.”
Gas hydrate: Refer to the definition of “hydrate.”
Gas-cap gas: The gas produced from a gas cap. See “associated gas.”
Gas condensate reservoir: Refer to definition of “condensate gas reservoir.”
© 2006 by Taylor and Francis Group, LLC
344 Fundamentals of Natural Gas Processing
Gravity, Baume: An arbitrary scale that expresses the relative density of
liquid products. For liquids lighter than water, the gravity Baume can be calculated
from the following equation: degrees Baume = (140/Sp. Gr.) − 130. For liquids
heavier than water, the formula is: degrees Baume = 145 − (145/Sp. Gr.).
*GTL: Gas to liquid; refers to plants designed to convert natural gas to
hydrocarbon liquids.
Heating value (heat of combustion): The amount of heat developed by the
complete combustion of a material. For natural gas, the heating value is usually
expressed as the gross or higher heating value in Btu per cu. ft. of gas at designated
conditions and either on the dry or water saturated basis. The gross, or higher, heating
value (normally referred to in the United States) is that measured in a calorimeter,
or computed from gas composition, when the heat of condensation of the water
produced is included in the total measured heat. The net, or lower, heating value
(normally referred to in Europe) is that obtained when the water produced by a
combustion process is not condensed, as is the usual circumstance in equipment that
burns fuel gas. The net value is the maximum portion of the heating value that can
be utilized in usual equipment. The difference between the gross and net heating
values is the heat that could be recovered if the water produced could be condensed.
High molecular weight absorption oil: Absorption oil that has a molecular
weight in excess of 180. Such oil is ordinarily used in nonrefrigerated absorption
plants that have distillation systems that employ stripping steam.
Hot carbonate process: A process for removing the bulk of the acid gases
from a gas stream by contacting the stream with a water solution of potassium
carbonate at a temperature in the range of 220°F to 240°F (104°C to 116°C).
Hydrate: A solid material that results from the combination of a hydrocarbon
with water under pressure.
Hydrate off: Stoppage of the flow of fluid by formation of hydrate crystals.
Injection gas: Gas injected into the producing formation to maintain or increase
the reservoir pressure or reduce the rate of decline of the reservoir pressure.
Injection well: The well through which the injection gas or other fluid flows
into the underground formation.
Iron sponge process: The method for removing small concentrations of
hydrogen sulfide from natural gas by passing the gas over a bed of wood shavings
that have been impregnated with a form of iron oxide. The impregnated wood
shavings are called “iron sponge.” The hydrogen sulfide reacts with the iron oxide
and forms iron sulfide and water. Regeneration, if desirable, may be accomplished
by exposing the depleted sponge bed to the oxygen in the air.
Jumbo tank cars: Tank cars that have capacities of 30,000 gallons (114
cubic meters) or more. Another group of cars known as “small jumbo” have
capacities that range from 18,000 to 22,000 gallons (68 to 83 cubic meters).
Various other sizes are also used. “Standard” tank cars have a capacity of 10,000
to 11,000 gallons (38 to 42 cubic meters).
Knockout (liquid): Any liquid condensed from a stream by a scrubber after
compression and cooling.
© 2006 by Taylor and Francis Group, LLC
Appendix A 345
Lead-acetate test: A method for detecting the presence of hydrogen sulfide in
a fluid by discoloration of paper that has been moistened with lead-acetate solution.
Lean amine: Amine solution that has been stripped of absorbed acid gases
and gives a solution suitable for recirculation to the contactor.
*LEAR (Lowest Achievable Emission Rate): Required on major new or
modified emission sources in nonattainment areas.
Lean gas: (1) The residue gas that remains after the recovery of natural gas
liquids in a gas processing plant. (2) Unprocessed gas that contains little or no
recoverable natural gas liquids.
Lean oil: Absorption oil purchased or made by the plant or oil from which
the absorbed constituents have been removed.
Lift gas: Gas injected into a well to assist in raising liquid to the surface.
Low molecular weight absorption oil: Absorption oil that has a molecular
weight below 155. Commonly used in refrigerated absorption plants that have
dry distillation systems.
LNG (liquefied natural gas): The light hydrocarbon portion of natural gas,
predominately methane, that has been liquefied.
*LPG (liquefied petroleum gas): Product streams that contain primarily
propane and butane, with minor amounts of ethane.
LTX unit (low temperature extraction unit): A unit that uses the refriger-
ating effect of the adiabatic expansion of a gas for improved liquid recovery from
streams that are produced from high pressure gas-condensate reservoirs. Some-
times called low temperature separators (LTS).
Mainline plant: A plant that processes the gas that is being transported
through a cross-country transmission line. Also called pipeline, on-line, or straddle
plant.
Make-up gas: (1) Gas that is taken in succeeding years that has been paid
for previously under a “take or pay” clause in a gas purchase contract. The contract
will normally specify the number of years after payment in which the purchaser
can take delivery of make-up gas without paying a second time. (2) Gas injected
into a reservoir to maintain a constant reservoir pressure and thereby prevent
retrograde condensation. (3) In gas processing, a reduction in gas volume occurs
because of plant loss (fuel and shrinkage). Some agreements between gas trans-
mission companies and plant owners require plant losses to be made up or paid
for. The same may apply to Btu reduction as well.
MCF (thousand cubic feet – 28.32 m
3
): A standard unit for measuring or
expressing the volume of a thousand cubic feet of gas. The pressure and temper-
ature conditions for the standard measurement must be defined.
MEA (monoethanolamine): An amine for treating gas. Refer to definition
MER: Has two general meanings, the first of which is the more common. (1)
Maximum efficient rate is the highest rate at which a well or reservoir may be
produced without causing physical waste in the reservoir. (2) Most efficient rate is
of “amine.”
See gas lift.
© 2006 by Taylor and Francis Group, LLC
346 Fundamentals of Natural Gas Processing
the highest rate at which a well or reservoir can be produced without either reservoir
or surface physical waste. For example, a reservoir may be produced at the maximum
efficient rate, but at such a rate, gas production will be in excess of the capacity of
facilities in the field to handle the gas, so a lower, or most efficient, rate is set up for
the reservoir to avoid surface waste of valuable hydrocarbons in the form of flared gas.
Natural gasoline: A mixture of hydrocarbons, mostly pentanes and heavier,
extracted from natural gas, which meets vapor pressure, endpoint, and other
specifications for natural gasoline as adopted by the GPA.
Natural gasoline plant: One of the terms, now obsolete, used to denote a
natural gas processing plant. Refer to definition of “gas processing” and “gas
processing plant.”
NGL (natural gas liquids): Natural gas liquids are those hydrocarbons
liquefied at the surface in field facilities or in gas processing plants. Natural gas
liquids include propane, butanes, and natural gasoline. The raw product stream
from a plant recovering C
2
+ is known as “Y-grade” NGL.
Odorant: A highly odiferous fluid or gas, usually a light mercaptan, added
to a gas or LP gas to impart to it a distinctive odor for safety precautions and to
facilitate detection of leaks.
Oil-well gas: Gas that is produced from an oil well.
Operating factor: The percentage of time that a unit is performing its
function; for example, if a unit runs 800 hours (on stream time), takes 100 hours
for reconditioning and inspection, and takes 100 hours for starting up and shutting
Outage: The difference between the full interior volume of a storage vessel
or sample container and the volume of liquid therein. For gasoline and lighter
products, the regulatory bodies set a minimum limit for outage to provide space
for expansion of the liquid.
Peak day requirements: Maximum requirement of gas for a 24-hour period.
The quantity may be considerably greater than the daily average flow.
Peak shaving: The use of fuel and equipment to generate or manufacture
gas to supplement the normal supply of pipeline gas during periods of extremely
high demand.
*Peak shaving plant: A liquefaction facility designed to produce LNG for
storage.
Pigging: A procedure of forcing a solid object through a pipeline for cleaning
purposes.
Pipeline gas: Gas that meets a transmission company’s minimum specifications.
*Pulsation dampeners: Internally baffled vessel attached to the suction and
outlet of reciprocating compressors to dampen the pressure pulsations.
Propane, HD-5: A special grade of propane that consists predominately of
propane and that conforms to the GPA specification for HD-5 propane.
*RACT (Reasonably Available Control Technology): Technology is usu-
ally required for emissions from a new or modified commercial emission source,
down, the operating factor is 80%. Refer to definition of “stream day.”
© 2006 by Taylor and Francis Group, LLC
Appendix A 347
such as a plant or combustion engine, in a clean area. RACT standards govern
all major sources that have the potential to emit 100 tons per year (tpy) of VOCs
and NO
X
.
Raw gas: Unprocessed gas or the inlet gas to a plant.
Raw mix liquids: A mixture of natural gas liquids before fractionation. Also
called “raw make.”
Reclaimer: A system in which undesirable high-boiling contaminants of a
stream are separated from the desired lighter materials; a purifying still.
Residue: The material that remains after a separation process. (1) Residue
gas is the gas that remains after the recovery of liquid products. (2) Residue may
also be the heaviest liquid or solid that remains after the laboratory distillation
or some reclaiming process.
Retrograde condensation: Condensation that occurs in a pressure region
when the pressure on the gas that contains methane and heavier hydrocarbons is
reduced. This process is the reverse of usual behavior at lower pressures. Hence,
the term “retrograde condensation” is used to describe this phenomenon.
Retrograde pressure region: The pressure region wherein the hydrocarbons
exhibit increased volatility as the pressure increases.
Rich amine: The amine that leaves the bottom of the contactor. It is the lean
amine plus the acid gases removed from the gas by the lean amine.
Rich gas: A gas that is suitable as feed to a gas processing plant and from
which products can be extracted.
Rich oil: The oil that leaves the bottom of the absorber. It is the lean oil plus
all of the absorbed products.
RVP (Reid vapor pressure): A vapor pressure specification for the heavier
liquid products as determined by ASTM test procedure D-323. The vapor pressure
is reported as pounds per sq. in. Reid. The pressure approximates, but is somewhat
less than, the absolute vapor pressure of the liquid.
*SCADA (Supervisory Control and Data Acquisition): Computer system
used to both control and monitor field and plant systems.
Shrinkage: (1) The reduction in volume, heating value, or both of a gas
stream caused by the removal of some of its constituents. (2) Sometimes referred
to as the unaccounted loss of products from storage tanks.
Slop, or slop oil: A term rather loosely used to denote a mixture of oil
produced at various places in the plant that must be rerun or further processed
to be suitable for use.
*Slug catcher: A separator that is designed to separate intermittent large
volumes of liquids from a gas stream. It may be vessels or a manifolded pipe
system.
Solution gas: Gas that originates from the liquid phase in the oil reservoir.
Sour: Liquids and gases are said to be “sour” if they contain hydrogen sulfide,
mercaptans, or both over a specified level.
*Snubber: See pulsation dampener.
© 2006 by Taylor and Francis Group, LLC
348 Fundamentals of Natural Gas Processing
Sour gas: Gas that contains an appreciable quantity of carbon dioxide, hydro-
gen sulfide, or mercaptans.
Splitter: A name applied to fractionators, particularly those that separate
*SPA: Sales and purchase agreement
Sponge absorption unit: The unit wherein the vapors of the lighter absorp-
tion oils are recovered.
SRU: Sulfur recovery unit
Stabilized condensate: Condensate that has been stabilized to a definite
vapor pressure in a fractionation system.
Stabilizer: A name for a fractionation system that stabilizes any liquid (i.e.,
reduces the vapor pressure so that the resulting liquid is less volatile).
Standard cubic feet (scf): This term refers to a gas volume measurement at
a specified temperature and pressure. The temperature and pressure may be
defined in the gas sales contract or by reference to other standards. The commonly
used temperature and pressure is 60°F and 14.696 psia.
Stranded reserve: A gas reserve located where (1) no economic use exists
for the gas at the point of origin, and (2) transportation of the gas by pipeline
from the reserve to its point of end use is not feasible.
Strapping: A term applied to the process of calibrating liquid storage capacity
of storage tanks by increments of depth.
Stream day: This terms refers to a basis for calculating plant production. A
stream day is a day of full operation. This concept is different from a calendar
day, which would be used to give average production for a full year.
Stripper: A column wherein absorbed constituents are stripped from the
absorption oil. The term is applicable to columns using a stripping medium, such
as steam or gas.
Stripping factor: An expression used to explain the degree of stripping.
Mathematically, it is KV/L, the reciprocal of the absorption factor.
Substitute natural gas (SNG): Refer to definition of “synthetic gas.”
Sweet: This term refers to the near or absolute absence of sulfur compounds
in either gas or liquid as defined by a given specification standard.
Sweet gas: Gas that has no more than the maximum sulfur content defined
by (1) the specifications for the sales gas from a plant or (2) the definition by a
legal body such as the Railroad Commission of Texas.
Synthetic gas (SG): The preferred term to describe the salable gas product
that results from the gasification of coal or gas liquids or heavier hydrocarbons.
Tail gas: The exit gas from a plant.
Tailgate: The point within the gas processing plant at which the residue gas
is last metered. This point is usually at the plant residue sales meter or the
allocation meter.
Straddle plant: Refer to definition of “mainline plant.”
isomers (i.e., butane splitter refers to a debutanizer).
© 2006 by Taylor and Francis Group, LLC
Appendix A 349
Take-or-pay clause: The clause that may be in a contract that guarantees
pay to the seller for a gas even though the particular gas volume is not taken
during a specified time period. Some contracts may contain a time period for the
buyer to take later delivery of the gas without penalty.
TGCU (tail gas cleanup unit): A process unit that takes tail gas from an
SRU and removes additional sulfur.
Therm: A unit of gross heating value equivalent to 100,000 Btu (1.055056 × 10
8
J).
Tonne: A unit of mass measurement, commonly used in international petro-
leum commerce; an expression for the metric ton, or 1,000 kilograms.
*Vane pack: An insert in gas-liquid separators, where gas goes through a
torturous path to force mists to coalesce into droplets and drop out of the gas
phase.
Vapor pressure gasoline: A descriptive phrase for natural gasoline that meets
a specified vapor pressure.
Vapor pressure, GPA: Vapor pressure as specified by GPA procedures.
Weathering: The evaporation of liquid by exposure to the conditions of
atmospheric temperature and pressure. Partial evaporation of liquid by use of heat
may also be called weathering.
Weathering test: A GPA test for LP gas for the determination of heavy
components in a sample by evaporation of the sample as specified.
Wet gas: (1) A gas that contains water, or a gas that has not been dehydrated.
(2) A term synonymous with rich gas; that is, a gas from which the products have
White oil: A term for oil that has no color.
Wild gasoline: An obsolete term for natural gas liquids before fractionation
and stabilization.
Wobbe number: A number proportional to the heat input to a burner at
constant pressure. In British practice, it is the heating value of a gas divided by
the square root of its gravity. Widely used in Europe, together with measured or
calculated flame speed, to determine the interchangeability of gases.
Ullage: See outage.
not been extracted. Refer to the definition of “rich gas.”
© 2006 by Taylor and Francis Group, LLC
351
Appendix B
Physical Constants and
Physical Properties
CONTENTS OF TABLES AND FIGURES
Table B.1 Physical Properties of Compounds Typically Seen in Gas
Processing....................................................................................356
Table B.2 Liquid Density Computed value has units of kmol/m
3
..............363
Table B.3 Vapor Pressure Computed value has units of Pa .......................364
Table B.4 Heat of Vaporization Computed value has units of J/kmol .......365
Table B.5 Liquid Heat Capacity Computed value has units of J/kmol-K..366
Table B.6 Ideal Gas Heat Capacity Computed value has units of
J/kmol-K......................................................................................367
Table B.7 Liquid Viscosity Computed value has units of Pa-s ..................368
Table B.8 Vapor Viscosity at Low Pressure Computed value has units of
Pa-s ..............................................................................................369
Table B.9 Liquid Thermal Conductivity Computed value has units of
W/m-K.........................................................................................370
Table B.10 Vapor Thermal Conductivity at Low Pressure Computed value
has units of W/m-K.....................................................................371
Table B.11 Surface Tension Computed value has units of N/m...................372
Table B.12 Equations Noted in Tables B.2 to B.11......................................373
Table B.13 Molar Ideal Gas Heat Capacities for Various Gases
(Btu/(lb-mol-°R)) ........................................................................374
Table B.14 Molar Ideal Gas Heat Capacities for Various Gases
[kJ/(kmol-K)] ..............................................................................375
Table B.15 Selected Physical Properties of Alkanolamines .........................376
Table B.16 Selected Properties of Glycols Used for Dehydration...............382
Table B.17 Selected Solubility for Compounds in Glycols at 25°C
(g solute/100 ml of glycol) .........................................................383
Table B.18 Vapor Pressure Equations for Glycols........................................383
Table B.19 Properties of Saturated Steam in Engineering Units .................390
Table B.20 Properties of Saturated Water in SI Units ..................................392
Table B.21 Properties of Saturated Methane in Engineering Units .............394
Table B.22 Properties of Saturated Methane in SI Units .............................395
© 2006 by Taylor and Francis Group, LLC
352 Fundamentals of Natural Gas Processing
Table B.23 Properties of Saturated Ethane in Engineering Units ................396
Table B.24 Properties of Saturated Ethane in SI Units ................................398
Table B.25 Properties of Saturated Ethylene in Engineering Units .............400
Table B.26 Properties of Saturated Ethylene in SI Units .............................402
Table B.27 Properties of Saturated Propane in Engineering Units ..............404
Table B.28 Properties of Saturated Propane in SI Units ..............................406
Figure B.1 Densities of monoethanolamine (MEA) – water mixtures as a
function of composition and temperature. (Adapted from Dow
Chemical, 2003b.)......................................................................376
Figure B.2 Densities of diethanolamine (DEA) – water mixtures as a function
of composition and temperature. (Adapted from Dow Chemical,
2003b.) .......................................................................................377
Figure B.3 Densities of triethanolamine (TEA) – water mixtures as a function
of composition and temperature. (Adapted from Dow Chemical,
2003b.) .......................................................................................377
Figure B.4 Densities of methyl diethanolamine (MDEA) – water mixtures as
a function of composition and temperature. (Adapted through the
courtesy of INEOS Oxide)........................................................378
Figure B.5 Heat capacities of monoethanolamine (MEA) – water mixtures as
a function of composition and temperature. (Adapted from Dow
Chemical, 2003b.)......................................................................378
Figure B.6 Heat capacities of diethanolamine (DEA) – water mixtures as a
function of composition and temperature. (Adapted from Dow
Chemical, 2003b.)......................................................................379
Figure B.7 Heat capacities of triethanolamine (TEA) – water mixtures as a
function of composition and temperature. (Adapted from Dow
Chemical, 2003b.)......................................................................379
Figure B.8 Heat capacities of methyldiethanolamine (MDEA) – water
mixtures as a function of composition and temperature. (Adapted
through the courtesy of INEOS Oxide). ...................................380
Figure B.9 Viscosities of monoethanolamine (MEA) – water mixtures as a
function of composition and temperature. (Adapted from Dow
Chemical, 2003b.)......................................................................380
Figure B.10 Viscosities of diethanolamine (DEA) – water mixtures as a
function of composition and temperature. (Adapted from Dow
Chemical, 2003b.) Reference implies that DEA and TEA
viscosities are sufficiently close that this figure can be used to
estimate the viscosity of TEA–water mixtures.........................381
Figure B.11 Viscosities of methyldiethanolamine (MDEA) – water mixtures as
a function of composition and temperature. (Adapted through the
courtesy of INEOS Oxide)........................................................381
Figure B.12 Specific gravities of ethylene glycol–water mixtures as a function
of concentration and temperature. (Courtesy of MEGlobal.)...384
Figure B.13 Specific gravities of diethylene glycol–water mixtures as a function
of concentration and temperature. (Courtesy of MEGlobal.)...384
© 2006 by Taylor and Francis Group, LLC
Appendix B 353
Figure B.14 Specific gravities of triethylene glycol–water mixtures as a
function of concentration and temperature. (Courtesy of Dow
Chemical.)..................................................................................385
Figure B.15 Specific gravities of tetraethylene glycol–water mixtures as a
function of concentration and temperature. (Courtesy of Dow
Chemical.)..................................................................................385
Figure B.16 Specific heats of ethylene glycol–water mixtures as a function of
concentration and temperature. (Courtesy of MEGlobal.) .......386
Figure B.17 Specific heats of diethylene glycol–water mixtures as a function
of concentration and temperature. (Courtesy of MEGlobal.)...386
Figure B.18 Specific heats of triethylene glycol–water mixtures as a function
of concentration and temperature. (Courtesy of Dow
Chemical.)..................................................................................387
Figure B.19 Specific heats of tetraethylene glycol–water mixtures as a function
of concentration and temperature. (Courtesy of Dow
Chemical.)..................................................................................387
Figure B.20 Viscosities of ethylene glycol–water mixtures as a function of
concentration and temperature. (Courtesy of MEGlobal.) .......388
Figure B.21 Viscosities of diethylene glycol–water mixtures as a function of
concentration and temperature. (Courtesy of MEGlobal.) .......388
Figure B.22 Viscosities of triethylene glycol–water mixtures as a function of
concentration and temperature. (Courtesy of Dow
Chemical.)..................................................................................389
Figure B.23 Viscosities of tetraethylene glycol–water mixtures as a function of
concentration and temperature. (Courtesy of Dow Chemical.) 389
Figure B.24 Pressure–enthalpy diagram for methane in engineering units
(ASHRAE, 2005). .....................................................................408
Figure B.25 Pressure–enthalpy diagram for methane in SI units
(ASHRAE, 2005). .....................................................................409
Figure B.26 Pressure–enthalpy diagram for ethane in engineering units
(ASHRAE, 2005). .....................................................................410
Figure B.27 Pressure–enthalpy diagram for ethane in SI units
(ASHRAE, 2005). .....................................................................411
Figure B.28 Pressure–enthalpy diagram for ethylene in engineering units
(ASHRAE, 2005). .....................................................................412
Figure B.29 Pressure–enthalpy diagram for ethylene in SI units
(ASHRAE, 2005). .....................................................................413
Figure B.30 Pressure–enthalpy diagram for propane in engineering units
(ASHRAE, 2005). .....................................................................414
Figure B.31 Pressure–enthalpy diagram for propane in SI units
(ASHRAE, 2005). .....................................................................415
Figure B.32 Pseudocritical temperature and pressure as a function of gas
specific gravity. Applicable to natural gas mixtures that contain
less than 5 mol% N
2
, 2 mol% CO
2
, and 2 mol% H
2
S. (Adapted
from Engineering Data Book, 1987.) .......................................416
© 2006 by Taylor and Francis Group, LLC
354 Fundamentals of Natural Gas Processing
Figure B.33 Compressibility factor chart as a function of pseudocritical
temperature and pressure. Chart applies to gases with a molar mass
of less than 40. (Adapted from Engineering Data
Book, 2004c.) ............................................................................417
B.1 UNIT CONVERSION FACTORS
Listed below are some of the more commonly used conversion factors.
Mass
1 pound (lb) 7,000 grains 16.0 ounces (oz)
453.5924 grams (g)
Length
1 foot (ft) 12.0 inches (in)
30.480 centimeters (cm) 0.30480 meters (m)
Temperature
°C (°F −32)/1.8
K °C + 273.15
K °R/1.8
°F 1.8(°C) + 32
°R °F + 459.67
°R 1.8(K)
Volume
1 cubic foot (ft
3
) 7.48052 gallons (gal) 1,728 cubic inches
0.1781076 barrels (42 U.S. gal) of oil (bbl)
28.31685 liters (L) 0.02831685 cubic meters (m
3
)
Density
1 lb
m
/ft
3
0.1336806 lb
m
/gal
0.01601846 g/cm
3
16.01846 kg/m
3
Force
1 pound (lb
f
) 4.448 × 10
5
dyne (dyn) 4.448222 Newtons (N)
Pressure
1 atmosphere (atm) 1.01325 bar 14.696 lb
f
/in
2
760 mm Hg (at 32°F)
1.013250 × 10
5
Pascal (Pa)
Energy
1 British Thermal Unit (Btu) (IT) 252.1644 cal (tc) 3.930148 × l0
−4
hp
1.055056 × 10
3
joules (J) 2.930711 × 10
−4
kWh
(1 Btu [IT] 1.00067 Btu [tc])
(Note: Customarily the Btu refers to the International Steam Table [IT] Btu,
and the calorie refers to the thermochemical calorie [tc])
Flow
1 gal/min 0.1336805 ft
3
/min 1.42857 bbl/h 6.309020 × 10
−5
m
3
/s
Power
1 hp (US) 2544.433 Btu (IT)/h 550 ft lb
f
/s 745.6999 watts (W)
Specific Energy per Degree
1 Btu/lb
m
-°F (IT) 1.0 cal/g-°C (IT) 4.186800 kJ/kg-K
© 2006 by Taylor and Francis Group, LLC
Appendix B 355
B.2 GAS CONSTANTS AND STANDARD GAS
CONDITIONS
B.2.1 UNIVERSAL GAS CONSTANTS
Metric/SI Units
8.31441 J/(mol-K)
1.98719 cal/(mol-K)
82.0568 cm
3
-atm/(mol-K)
62.3633 L-mm/(mol-K)
0.0820568 L-atm/(mol-K)
8.31441 m
3
-Pa/(mol-K)
8.31441 × 10
−5
m
3
-bar/(mol-K)
0.0831441 L-bar/(mol-K)
Engineering and Mixed Units
1.98585 Btu/(lb-mol-°R)
0.730235 atm-ft
3
/(lb-mol-°R)
0.739911 bar-ft
3
/(lb-mol-°R)
1545.34 ft-lb
f
/(lb-mol-°R)
10.7315 psi ft
3
/(lb-mol-°R)
B.2.2 STANDARD GAS CONDITIONS

Gas volumes given are for ideal gas only.
1. Normal: continental and scientific applications
0°C, 1.01325 bar, gas volume 22.4136 m
3
/kg-mol
32°F, 14.696 psia, gas volume 359.031 ft
3
/lb-mol
(Nm
3
based upon these conditions)
2. Standard (scf): U.S. engineering applications
15.5556°C, 1.01325 bar, gas volume 23.6900 m
3
/kg-mol
60°F, 14.696 psia, gas volume 379.49 ft
3
/lb-mol
(scf standard cubic foot based upon these conditions)
3. Standard: API 2564, SI 15°C, 101.325 kPa, gas volume 23.6444
m
3
/kg-mol (Sm
3
based upon these conditions)
B.3 THERMODYNAMIC AND PHYSICAL
PROPERTY DATA
B.3.1 TABLE OF PHYSICAL CONSTANTS OF PURE FLUIDS
tered in natural gas processing. For additional properties the reader should consult
GPA Standard 2145-03 (2005). The standard includes a number of notes regarding
the data and data sources.
Table B.1 provides some of the most useful physical constants for fluids encoun-
3
5
6
F
u
n
d
a
m
e
n
t
a
l
s

o
f

N
a
t
u
r
a
l

G
a
s

P
r
o
c
e
s
s
i
n
g
TABLE B.1
Physical Properties of Compounds Typically Seen in Gas Processing
Component Methane Ethane Propane Isobutane n-Butane Isopentane n-Pentane
Molar mass 16.042 30.069 44.096 58.122 58.122 72.149 72.149
Boiling Point, °F (K) −258.67 (111.67) −127.48 (184.55) −43.72 (231.08) 11.08 (261.53) 31.09 (272.64) 82.11 (300.99) 96.98 (309.25)
Freezing Point ,°F (K) −296.45 (90.68) −297.04 (90.35) −305.73 (85.52) −255.3 (113.54) −217.05 (134.79) −255.8 (113.26) −201.5 (143.43)
Vapor Pressure at 100 °F
(313.15K), psia (bar)
5000 (350)
a
800 (55)
a
188.69 (13.699) 72.484 (5.3012) 51.683 (3.7936) 20.456 (1.5136) 15.558 (1.1556)
Density of liquid
Relative to water 0.3
a
0.35643 0.50738 0.56295 0.58408 0.6246 0.63113
Density, lb/gal (kg/m
3
) 2.5 (300)
a
2.9716 (358) 4.2301 (507.67) 4.6934 (563.07) 4.8696 (584.14) 5.2074 (624.54) 5.2618 (631.05)
Density of ideal gas
Relative density, air 1.0 0.55397 1.0383 1.5227 2.0071 2.0071 2.4914 2.4914
Density lb/1000 ft
3
(kg/m
3
) 42.274 (0.67848) 79.237 (1.2717) 116.2 (1.8649) 153.16 (2.4581) 153.16 (2.4581) 190.12 (3.0514) 190.12 (3.0514)
Volume of liquid
gal/lb-mol (cm
3
/mol) 6.417 (53.475)
a
10.119 (83.992) 10.424 (86.859) 12.384 (103.22) 11.936 (99.501) 13.855 (115.52) 13.712 (114.33)
Critical conditions
Temperature, °F (K) −116.66 (190.56) 89.924 (305.33) 205.92 (369.77) 274.41 (407.82) 305.546 (425.12) 368.98 (460.36) 385.75 (469.68)
Pressure, psia (bar) 667 (45.99) 706.6 (48.72) 615.5 (42.44) 527.9 (36.40) 550.9 (37.98) 490.4 (33.81) 488.8 (33.70)
© 2006 by Taylor and Francis Group, LLC
A
p
p
e
n
d
i
x

B
3
5
7
Gross heating value, ideal reaction
Fuel as liquid Btu/lb,
(MJ/kg)
22181 (51.594) 21490 (49.988) 21080 (49.033) 21136 (49.165) 20891 (48.594) 20923 (48.669)
Fuel as liquid, Btu/gal
(MJ/m
3
)
65914 (18471) 90905 (25377) 98935 (27609) 102926 (28719) 108789 (30349) 110094 (30712)
fuel as ideal gas, Btu/lb
(MJ/kg)
23892 (55.576) 22334 (51.952) 21654 (50.37) 21232 (49.389) 21300 (49.547) 21044 (48.95) 21085 (49.046)
Fuel as ideal gas Btu/ft
3

(MJ/m
3
)
1010 (37.707) 1769.7 (66.067) 2516.2 (93.936) 3252 (121.4) 3262.4 (121.79) 4000.9 (149.36) 4008.7 (149.66)
Btu/gal, fuel as ideal gas 59730 66369 91599 99652 103724 109584 110946
Net heating value, ideal reaction, fuel as ideal gas
Btu/ft
3
, (MJ/m
3
) 909 (33.949) 1619 (60.429) 2315 (86.419) 3000 (112.01) 3011 (112.4) 3699 (138.09) 3707 (138.38)
Heat of vaporization at 14.696 psia (1.01325 bar)
Btu/lb, kJ/kg 219.8 (511.3) 210.3 (489.2) 183.4 (426.7) 157.2 (365.6) 166.3 (386.9) 147.7 (343.7) 154.4 (359.2)
Flammability limits at 100 °F, 14.696 psia (313.15K, 1.01325 bar) in air, vol%
Lower limit 5 2.9 2 1.8 1.5 1.3 1.4
Upper limit 15 13 9.5 8.5 9 8 8.3
a
hypothetical value because liquid does not exist at stated condition
(continued)
© 2006 by Taylor and Francis Group, LLC
3
5
8
F
u
n
d
a
m
e
n
t
a
l
s

o
f

N
a
t
u
r
a
l

G
a
s

P
r
o
c
e
s
s
i
n
g
TABLE B.1 (Continued)
Component n-Hexane n-Heptane n-Octane n-Nonane n-Decane Carbon Dioxide
Molar mass 86.175 100.202 114.229 128.255 142.282 44.01
Boiling point, °F (K) 155.72 (341.88) 209.13 (371.56) 258.21 (398.82) 303.4 (423.93) 345.4 (447.26) −109.12 (194.75)
Vapor pressure at 100°F
(313.15K), psia (bar)
4.961 (37.3) 1.62 (12.336) 0.5366 (4.1403) 0.17 (1.3488) 0.0616 (0.4876)
Density of liquid
Relative to water 0.66405 (0.66448) 0.68819 (0.6886) 0.70698 (0.70737) 0.72186 (0.72224) 0.73406 (0.73442) 0.82203 (0.82195)
Density, lb/gal (kg/m
3
) 5.5363 (663.89) 5.7375 (687.98) 5.8942 (706.73) 6.0183 (721.59) 6.12 (733.76) 6.8534 (821.22)
Density of Ideal Gas
Relative density, air 1.0 2.9758 3.4601 3.9445 4.4289 4.9132 1.5197
Density lb/1,000 ft
3
(kg/m
3
) 227.09 (3.6446) 264.05 (4.2378) 301.01 (4.831) 337.97 (5.4242) 374.93 (6.0174) 115.97 (1.8613)
Volume of liquid
gal/lb-mol (cm
3
/mol) 15.566 (129.8) 17.464 (145.65) 19.38 (161.63) 21.311 (177.74) 23.249 (193.91) 6.4216 (53.59)
Critical conditions
Temperature, °F (K) 453.83 (507.5) 512.87 (540.3) 564.22 (568.8) 610.8 (594.7) 652.2 (617.7) 87.8 (304.1)
Pressure, psia (bar) 436.9 (3012) 396.8 (2736) 360.7 (2487) 330.7 (2280) 304.6 (2100) 1070 (7377)
© 2006 by Taylor and Francis Group, LLC
A
p
p
e
n
d
i
x

B
3
5
9
Gross heating value, ideal reaction
Fuel as liquid Btu/lb, (MJ/kg) 20783 (48.343) 20680 (48.104) 20601 (47.92) 20543 (47.785) 20494 (47.671)
Fuel as liquid, Btu/gal (MJ/m
3
) 115060 (32094) 118654 (33095) 121428 (33866) 123634 (34481) 125424 (34979)
fuel as ideal gas, Btu/lb (MJ/kg) 20944 (48.717) 20839 (48.474) 20760 (48.289) 20701 (48.153) 20652 (48.038)
Fuel as ideal gas Btu/ft
3
(MJ/m
3
) 4756 (177.55) 5502.5 (205.42) 6248.9 (233.29) 6996.4 (261.19) 7743 (289.06)
Btu/gal, fuel as ideal gas 115951 119565 122363 124585 126388
Net heating value, ideal reaction, fuel as ideal gas
Btu/ft
3
, (MJ/m
3
) 4404 (164.4) 5100 (190.39) 5796 (216.37) 6493 (242.4) 7190 (268.39)
Heat of vaporization at 14.696 psia (1.01325 bar)
Btu/lbm, kJ/kg 144.1 (335.1) 136.7 (318.1) 130 (302.4) 124.4 (289.3) 119.3 (277.6) 246.5 (573.3)
Flammability limits at 100°F, 14.696 psia (313.15 K, 1.01325 bar) in air, vol%
Lower limit 1.1 1 0.8 0.7 0.7
Upper limit 7.7 7 6.5 5.6 5.4
(continued)
© 2006 by Taylor and Francis Group, LLC
3
6
0
F
u
n
d
a
m
e
n
t
a
l
s

o
f

N
a
t
u
r
a
l

G
a
s

P
r
o
c
e
s
s
i
n
g
TABLE B.1 (Continued)
Component Hydrogen Sulfide Nitrogen Oxygen Helium Air Water
Molar mass 34.082 28.013 31.999 4.0026 28.959 18.0153
Boiling Point, °F (K) −76.52 (212.86) −320.43 (77.355) −297.33
(90.188)
−452.06 (4.23) −317.64 (78.903) 211.95
(373.124)
Freezing Point ,°F (K) −121.81 (187.7) −346 (63.151) −361.82
(54.361)
−455.75 (2.177) 32
(273.15)
Vapor Pressure at 100 °F
(313.15K), psia (bar)
395.55 (28.67) 0.95051
(0.073849)
Density of liquid
Relative to water 0.80269 (0.80015) 0.80687 (0.8068) 1.1423 (1.1422) 0.87603
(0.87596)
1.0
Density, lb/gal (kg/m
3
) 6.6922 (799.4) 6.727 (806.1) 9.5236 (1141.2) 7.3036 (875.2) 8.3372
(999.103)
Density of gas
Relative density, air 1.0 1.1769 0.9673 1.105 0.1382 1.0 0.6221
Density lb/1000 ft
3
(kg/m
3
) 89.811 (1.4414) 73.819 (1.1847) 84.322 (1.3533) 10.547 (0.1693) 76.311 (1.2247) 47.473
(0.76191)
Volume of liquid
gal/lb-mol (cm
3
/mol) 5.0928 (42.63) 4.1643 (34.752) 3.36 (28.04) 3.965 (33.09) 2.1608
(18.031)
© 2006 by Taylor and Francis Group, LLC
A
p
p
e
n
d
i
x

B
3
6
1
Critical conditions
Temperature, °F (K) 212.8 (373.6) −232.53 (126.19) −181.43
(154.58)
−450.32 (5.195) −220.97 (132.61) 705.1
(647.1)
Pressure, psia (bar) 1306.5 (90.08) 492.5 (33.96) 731.4 (50.43) 33.0 (2.275) 551.9 (38.05) 3200.1 (22.064)
Gross heating value, ideal reaction
Fuel as liquid Btu/lb,
(MJ/kg)
6897 (12825)
Fuel as liquid, Btu/gal
(MJ/m
3
)
46156 (16.042)
Fuel as ideal gas, Btu/lb
(MJ/kg)
7093.8 (16.501)
Fuel as ideal gas Btu/ft
3

(MJ/m
3
)
637.11 (23.785)
Btu/gal, fuel as ideal gas 47473
Net heating value, ideal reaction, fuel as ideal gas
Btu/ft
3
, (MJ/m
3
) 586.8 (21.91)
Heat of vaporization at 14.696 psia (1.01325 bar)
Btu/lb, kJ/kg 234.4 (545.3) 85.63 (199.2) 91.6 (213.1) 8.922 (20.75) 86.81 (201.9) 970.12 (2256.5)
Flammability limits at 100°F, 14.696 psia (313.15K, 1.01325 bar) in air, vol%
Lower limit 4.3
Upper limit 45.5
Unless stated otherwise, pressure-dependent properties are given at 14.696 psia (1.01325 bar); temperature-dependent properties are given at 60°F (15°C); temperature-
dependent and pressure-dependent properties are given at 60°F (15°C) and 14.696 psia (1.01325 bar). Source: GPA (2005).
© 2006 by Taylor and Francis Group, LLC
© 2006 by Taylor and Francis Group, LLC
362 Fundamentals of Natural Gas Processing
B.3.2 TEMPERATURE-DEPENDENT PROPERTIES OF PURE FLUIDS
properties as a function of temperature. Pressure effect is not considered and
liquid properties are given at the vapor pressure of the fluid (saturation
boundary). The tables include the applicable temperature range and the esti-
mated quality of the prediction. They also include equation numbers for which
available by courtesy of the Design Institute for Physical Properties
(DIPPR,2005), an Industry Technology Alliance of the American Institute of
Chemical Engineers (AIChE), from the DIPPR 801 database, 2005 edition
reviewed data. The tables present 10 of the 15 temperature-dependent phys-
ical properties and 33 physical constants in the database. Nearly 1,800 com-
pounds are included in the database.

Tables B.2 to B.11 provide constants for equations to calculate physical
the coefficients apply. Table B.12 lists the equations. These data were made
(www.aiche.org/DIPPR). The coefficients were regressed from critically
A
p
p
e
n
d
i
x

B
3
6
3
TABLE B.2
Liquid Density (Computed value has units of kmol/m
3
)
Compound Equation Number Temperature Range (K) Quality A B C D E
Methane 105 90.7 − 190.6 < 1% 2.9214E+00 2.8976E−01 1.9056E+02 2.8881E−01
Ethylene 105 104 − 282.3 < 1% 2.0961E+00 2.7657E−01 2.8234E+02 2.9147E−01
Ethane 105 90.4 − 305.3 < 1% 1.9122E+00 2.7937E−01 3.0532E+02 2.9187E−01
Propane 105 85.5 − 369.8 < 1% 1.3757E+00 2.7453E−01 3.6983E+02 2.9359E−01
Isobutane 105 113.5 − 407.8 < 1% 1.0631E+00 2.7506E−01 4.078E+02 2.758E−01
n-Butane 105 134.9 − 425.1 < 1% 1.0677E+00 2.7188E−01 4.2512E+02 2.8688E−01
Isopentane 105 113.2 − 460.4 < 1% 9.1991E−01 2.7815E−01 4.604E+02 2.8667E−01
n-Pentane 105 143.4 − 469.7 < 1% 8.4947E−01 2.6726E−01 4.697E+02 2.7789E−01
Benzene 105 278.7 − 562 < 3% 1.0259E+00 2.6666E−01 5.6205E+02 2.8394E−01
n-Hexane 105 177.8 − 507.6 < 1% 7.0824E−01 2.6411E−01 5.076E+02 2.7537E−01
n-Heptane 105 182.6 − 540.2 < 1% 6.1259E−01 2.6211E−01 5.402E+02 2.8141E−01
Ammonia 105 195.4 − 405.6 < 1% 3.5383E+00 2.5443E−01 4.0565E+02 2.888E−01
Water 116 273.2 − 647.1 < 1% 1.7863E+01 5.8606E+01 −9.5396E+01 2.1389E+02 −1.4126E+02
Oxygen 105 54.4 − 154.6 < 1% 3.9143E+00 2.8772E−01 1.5458E+02 2.924E−01
Nitrogen 105 63.2 − 126.2 < 1% 3.2091E+00 2.861E−01 1.262E+02 2.966E−01
Hydrogen 105 14 − 33.2 < 3% 5.414E+00 3.4893E−01 3.319E+01 2.706E−01
Hydrogen sulfide 105 187.7 − 373.5 < 1% 2.7672E+00 2.7369E−01 3.7353E+02 2.9015E−01
Carbon monoxide 105 68.2 − 132.9 < 1% 2.897E+00 2.7532E−01 1.3292E+02 2.813E−01
Carbon dioxide 105 216.6 − 304.2 < 1% 2.768E+00 2.6212E−01 3.0421E+02 2.908E−01
Helium 105 2.2 − 5.2 < 3% 7.2475E+00 4.1865E−01 5.2E+00 2.4096E−01
Methanol 105 175.5 − 512.5 < 1% 2.3267E+00 2.7073E−01 5.125E+02 2.4713E−01
Ethylene glycol 105 260.2 − 720 < 1% 1.315E+00 2.5125E−01 7.2E+02 2.1868E−01
Diethylene glycol 105 262.7 − 744.6 < 3% 8.3692E−01 2.6112E−01 7.446E+02 2.422E−01
Triethylene glycol 105 266 − 769.5 < 3% 5.9672E−01 2.6217E−01 7.695E+02 2.4631E−01
Copyright 2005 by the Design Institute for Physical Properties (DIPPR), American Institute of Chemical Engineers (AIChE), and reproduced by permission of AIChE.
© 2006 by Taylor and Francis Group, LLC
3
6
4
F
u
n
d
a
m
e
n
t
a
l
s

o
f

N
a
t
u
r
a
l

G
a
s

P
r
o
c
e
s
s
i
n
g
TABLE B.3
Vapor Pressure (Computed value has units of Pa)
Compound Equation Number Temperature Range (K) Quality A B C D E
Methane 101 90.7 − 190.6 < 1% 3.9205E+01 −1.3244E+03 −3.4366E+00 3.1019E−05 2
Ethylene 101 104 − 282.3 < 1% 5.3963E+01 −2.443E+03 −5.5643E+00 1.9079E−05 2
Ethane 101 90.4 − 305.3 < 1% 5.1857E+01 −2.5987E+03 −5.1283E+00 1.4913E−05 2
Propane 101 85.5 − 369.8 < 3% 5.9078E+01 −3.4926E+03 −6.0669E+00 1.0919E−05 2
Isobutane 101 113.5 − 407.8 < 3% 1.0843E+02 −5.0399E+03 −1.5012E+01 2.2725E−02 1
n-Butane 101 134.9 − 425.1 < 3% 6.6343E+01 −4.3632E+03 −7.046E+00 9.4509E−06 2
Isopentane 101 113.2 − 460.4 < 3% 7.1308E+01 −4.976E+03 −7.7169E+00 8.7271E−06 2
n-Pentane 101 143.4 − 469.7 < 3% 7.8741E+01 −5.4203E+03 −8.8253E+00 9.6171E−06 2
Benzene 101 278.7 − 562 < 1% 8.3107E+01 −6.4862E+03 −9.2194E+00 6.9844E−06 2
n-Hexane 101 177.8 − 507.6 < 3% 1.0465E+02 −6.9955E+03 −1.2702E+01 1.2381E−05 2
n-Heptane 101 182.6 − 540.2 < 3% 8.7829E+01 −6.9964E+03 −9.8802E+00 7.2099E−06 2
Ammonia 101 195.4 − 405.6 < 1% 9.0483E+01 −4.6697E+03 −1.1607E+01 1.7194E−02 1
Water 101 273.2 − 647.1 < 0.2% 7.3649E+01 −7.2582E+03 −7.3037E+00 4.1653E−06 2
Oxygen 101 54.4 − 154.6 < 1% 5.1245E+01 −1.2002E+03 −6.4361E+00 2.8405E−02 1
Nitrogen 101 63.2 − 126.2 < 1% 5.8282E+01 −1.0841E+03 −8.3144E+00 4.4127E−02 1
Hydrogen 101 14 − 33.2 < 3% 1.269E+01 −9.4896E+01 1.1125E+00 3.2915E−04 2
Hydrogen sulfide 101 187.7 − 373.5 < 3% 8.5584E+01 −3.8399E+03 −1.1199E+01 1.8848E−02 1
Carbon monoxide 101 68.2 − 132.9 < 1% 4.5698E+01 −1.0766E+03 −4.8814E+00 7.5673E−05 2
Carbon dioxide 101 216.6 − 304.2 < 1% 1.4054E+02 −4.735E+03 −2.1268E+01 4.0909E−02 1
Helium 101 1.8 − 5.2 < 1% 1.1533E+01 −8.99E+00 6.724E−01 2.743E−01 1
Methanol 101 175.5 − 512.5 < 1% 8.2718E+01 −6.9045E+03 −8.8622E+00 7.4664E−06 2
Ethylene glycol 101 260.2 − 720 < 3% 8.409E+01 −1.0411E+04 −8.1976E+00 1.6536E−18 6
Diethylene glycol 101 262.7 − 744.6 < 10% 1.4245E+02 −1.505E+04 −1.6318E+01 5.9506E−18 6
Triethylene Glycol 101 266 − 769.5 < 10% 1.5248E+02 −1.6449E+04 −1.767E+01 6.4481E−18 6
Copyright 2005 by the Design Institute for Physical Properties (DIPPR), American Institute of Chemical Engineers (AIChE), and reproduced by permission of AIChE.
© 2006 by Taylor and Francis Group, LLC
A
p
p
e
n
d
i
x

B
3
6
5
TABLE B.4
Heat of Vaporization (Computed value has units of J/kmol)
Compound Equation Number Temperature Range(K) Quality A B C D
Methane 106 90.7 − 190.6 < 1% 1.0194E+07 2.6087E−01 −1.4694E−01 2.2154E−01
Ethylene 106 104 − 282.3 < 1% 1.8844E+07 3.6485E−01
Ethane 106 90.4 − 305.3 < 1% 2.1091E+07 6.0646E−01 −5.5492E−01 3.2799E−01
Propane 106 85.5 − 369.8 < 3% 2.9209E+07 7.8237E−01 −7.7319E−01 3.9246E−01
Isobutane 106 113.5 − 407.8 < 3% 3.188E+07 3.9006E−01
n-Butane 106 134.9 − 425.1 < 3% 3.6238E+07 8.337E−01 −8.2274E−01 3.9613E−01
Isopentane 106 113.2 − 460.4 < 3% 3.7593E+07 3.9173E−01
n-Pentane 106 143.4 − 469.7 < 3% 3.9109E+07 3.8681E−01
Benzene 106 278.7 − 562 < 1% 4.5346E+07 3.9053E−01
n-Hexane 106 177.8 − 507.6 < 3% 4.4544E+07 3.9002E−01
n-Heptane 106 182.6 − 540.2 < 3% 5.0014E+07 3.8795E−01
Ammonia 106 195.4 − 405.6 < 1% 3.1523E+07 3.914E−01 −2.289E−01 2.309E−01
Water 106 273.2 − 647.1 < 1% 5.2053E+07 3.199E−01 −2.12E−01 2.5795E−01
Oxygen 106 54.4 − 154.6 < 1% 9.008E+06 4.542E−01 −4.096E−01 3.183E−01
Nitrogen 106 63.2 − 126.2 < 1% 7.4905E+06 4.0406E−01 −3.17E−01 2.7343E−01
Hydrogen 106 14 − 33.2 < 3% 1.0127E+06 6.98E−01 −1.817E+00 1.447E+00
Hydrogen sulfide 106 187.7 − 373.5 < 3% 2.5676E+07 3.7358E−01
Carbon monoxide 106 68.1 − 132.5 < 1% 8.585E+06 4.921E−01 −3.26E−01 2.231E−01
Carbon dioxide 106 216.6 − 304.2 < 1% 2.173E+07 3.82E−01 −4.339E−01 4.2213E−01
Helium 106 2.2 − 5.2 < 10% 1.2504E+05 1.3038E+00 −2.6954E+00 1.7098E+00
Methanol 106 175.5 − 512.5 < 3% 5.0451E+07 3.3594E−01
Ethylene glycol 106 260.2 − 720 < 3% 8.3518E+07 4.2625E−01
Diethylene glycol 106 262.7 − 744.6 < 10% 1.0829E+08 5.4022E−01
Triethylene Glycol 106 266 − 769.5 < 10% 1.2127E+08 5.8261E−01
Copyright 2005 by the Design Institute for Physical Properties (DIPPR), American Institute of Chemical Engineers (AIChE), and reproduced by permission of AIChE.
© 2006 by Taylor and Francis Group, LLC
3
6
6
F
u
n
d
a
m
e
n
t
a
l
s

o
f

N
a
t
u
r
a
l

G
a
s

P
r
o
c
e
s
s
i
n
g
TABLE B.5
Liquid Heat Capacity (Computed value has units of J/kmol-K)
Compound Equation Number Temperature Range (K) Quality A B C D E
Methane 114 90.7 − 190 < 1% 6.5708E+01 3.8883E+04 −2.5795E+02 6.1407E+02
Ethylene 100 104 − 252.7 < 3% 2.4739E+05 −4.428E+03 4.0936E+01 −1.697E−01 2.6816E−04
Ethane 114 92 − 290 < 1% 4.4009E+01 8.9718E+04 9.1877E+02 −1.886E+03
Propane 114 85.5 − 360 < 1% 6.2983E+01 1.1363E+05 6.3321E+02 −8.7346E+02
Isobutane 100 113.5 − 380 < 3% 1.7237E+05 −1.7839E+03 1.4759E+01 −4.7909E−02 5.805E−05
n-Butane 100 134.9 − 400 < 1% 1.9103E+05 −1.675E+03 1.25E+01 −3.874E−02 4.6121E−05
Isopentane 100 113.2 − 310 < 3% 1.083E+05 1.46E+02 −2.92E−01 1.51E−03
n-Pentane 100 143.4 − 390 < 1% 1.5908E+05 −2.705E+02 9.9537E−01
Benzene 100 278.7 − 353.2 < 3% 1.2944E+05 −1.695E+02 6.4781E−01
n-Hexane 100 177.8 − 460 < 1% 1.7212E+05 −1.8378E+02 8.8734E−01
n-Heptane 114 182.6 − 520 < 1% 6.126E+01 3.1441E+05 1.8246E+03 −2.5479E+03
Ammonia 114 203.2 − 401.2 < 3% 6.1289E+01 8.0925E+04 7.994E+02 −2.651E+03
Water 100 273.2 − 533.2 < 1% 2.7637E+05 −2.0901E+03 8.125E+00 −1.4116E−02 9.3701E−06
Oxygen 100 54.4 − 142 < 3% 1.7543E+05 −6.1523E+03 1.1392E+02 −9.2382E−01 2.7963E−03
Nitrogen 100 63.2 − 112 < 3% 2.8197E+05 −1.2281E+04 2.48E+02 −2.2182E+00 7.4902E−03
Hydrogen 114 14 − 32 < 5% 6.6653E+01 6.7659E+03 −1.2363E+02 4.7827E+02
Hydrogen sulfide 114 187.7 − 370 < 3% 6.4666E+01 4.9354E+04 2.2493E+01 −1.623E+03
Carbon monoxide 114 68.2 − 132 < 3% 6.5429E+01 2.8723E+04 −8.4739E+02 1.9596E+03
Carbon dioxide 100 220 − 290 < 3% −8.3043E+06 1.0437E+05 −4.3333E+02 6.0052E−01
Helium 100 2.2 − 4.6 < 5% 3.8722E+05 −4.6557E+05 2.118E+05 −4.2494E+04 3.2129E+03
Methanol 100 175.5 − 400 < 1% 1.058E+05 −3.6223E+02 9.379E−01
Ethylene glycol 100 260.2 − 493.2 < 1% 3.554E+04 4.3678E+02 −1.8486E−01
Diethylene glycol 100 262.7 − 451.2 < 3% 1.2541E+05 4.0058E+02
Triethylene Glycol 100 265.8 − 441 < 3% 1.538E+05 5.87E+02
Copyright 2005 by the Design Institute for Physical Properties (DIPPR), American Institute of Chemical Engineers (AIChE), and reproduced by permission of AIChE.
© 2006 by Taylor and Francis Group, LLC
A
p
p
e
n
d
i
x

B
3
6
7
TABLE B.6
Ideal Gas Heat Capacity (Computed value has units of J/kmol-k)
Compound Equation Number Temperature Range (K) Quality A B C D E
Methane 107 50 − 1500 < 1% 3.3298E+04 7.9933E+04 2.0869E+03 4.1602E+04 9.9196E+02
Ethylene 107 60 − 1500 < 1% 3.338E+04 9.479E+04 1.596E+03 5.51E+04 7.408E+02
Ethane 107 200 − 1500 < 1% 4.0326E+04 1.3422E+05 1.6555E+03 7.3223E+04 7.5287E+02
Propane 107 200 − 1500 < 1% 5.192E+04 1.9245E+05 1.6265E+03 1.168E+05 7.236E+02
Isobutane 107 200 − 1500 < 1% 6.549E+04 2.4776E+05 1.587E+03 1.575E+05 7.0699E+02
n-Butane 107 200 − 1500 < 1% 7.134E+04 2.43E+05 1.63E+03 1.5033E+05 7.3042E+02
Isopentane 107 200 − 1500 < 1% 7.46E+04 3.265E+05 1.545E+03 1.923E+05 6.667E+02
n-Pentane 107 200 − 1500 < 1% 8.805E+04 3.011E+05 1.6502E+03 1.892E+05 7.476E+02
Benzene 107 200 − 1500 < 1% 4.4767E+04 2.3085E+05 1.4792E+03 1.6836E+05 6.7766E+02
n-Hexane 107 200 − 1500 < 1% 1.044E+05 3.523E+05 1.6946E+03 2.369E+05 7.616E+02
n-Heptane 107 200 − 1500 < 1% 1.2015E+05 4.001E+05 1.6766E+03 2.74E+05 7.564E+02
Ammonia 107 100 − 1500 < 1% 3.3427E+04 4.898E+04 2.036E+03 2.256E+04 8.82E+02
Water 107 100 − 2273.2 < 1% 3.3363E+04 2.679E+04 2.6105E+03 8.896E+03 1.169E+03
Oxygen 107 50 − 1500 < 1% 2.9103E+04 1.004E+04 2.5265E+03 9.356E+03 1.1538E+03
Nitrogen 107 50 − 1500 < 1% 2.9105E+04 8.6149E+03 1.7016E+03 1.0347E+02 9.0979E+02
Hydrogen 107 250 − 1500 < 1% 2.7617E+04 9.56E+03 2.466E+03 3.76E+03 5.676E+02
Hydrogen sulfide 107 100 − 1500 < 1% 3.3288E+04 2.6086E+04 9.134E+02 −1.7979E+04 9.494E+02
Carbon monoxide 107 60 − 1500 < 1% 2.9108E+04 8.773E+03 3.0851E+03 8.4553E+03 1.5382E+03
Carbon dioxide 107 50 − 5000 < 1% 2.937E+04 3.454E+04 1.428E+03 2.64E+04 5.88E+02
Helium 100 100 − 1500 < 1% 2.0786E+04
Methanol 107 200 − 1500 < 1% 3.9252E+04 8.79E+04 1.9165E+03 5.3654E+04 8.967E+02
Ethylene glycol 107 300 − 1500 < 3% 6.3012E+04 1.4584E+05 1.673E+03 9.7296E+04 7.7365E+02
Diethylene glycol 107 200 − 1500 < 10% 8.79E+04 2.713E+05 1.3963E+03 1.7035E+05 6.2404E+02
Triethylene Glycol 107 300 − 1500 < 25% 9.04E+04 4.202E+05 1.2628E+03 2.7705E+05 5.311E+02
Copyright 2005 by the Design Institute for Physical Properties (DIPPR), American Institute of Chemical Engineers (AIChE), and reproduced by permission of AIChE.
© 2006 by Taylor and Francis Group, LLC
3
6
8
F
u
n
d
a
m
e
n
t
a
l
s

o
f

N
a
t
u
r
a
l

G
a
s

P
r
o
c
e
s
s
i
n
g
TABLE B.7
Liquid Viscosity (Computed value has units of Pa-s)
Compound Equation Number Temperature Range (K) Quality A B C D E
Methane 101 90.7 − 188 < 3% −6.1572E+00 1.7815E+02 −9.5239E−01 −9.0606E−24 10
Ethylene 101 104 − 250 < 5% 1.8878E+00 7.8865E+01 −2.1554E+00
Ethane 101 90.4 − 300 < 3% −7.0046E+00 2.7638E+02 −6.087E−01 −3.1108E−18 7
Propane 101 85.5 − 360 < 5% −1.7156E+01 6.4625E+02 1.1101E+00 −7.3439E−11 4
Isobutane 101 110 − 311 < 5% −1.3912E+01 7.9709E+02 4.5308E−01
n-Butane 101 134.9 − 420 < 3% −7.2471E+00 5.3482E+02 −5.7469E−01 −4.6625E−27 10
Isopentane 101 150 − 310 < 1% −1.2596E+01 8.8911E+02 2.0469E−01
n-Pentane 101 143.4 − 465.2 < 3% −5.3509E+01 1.8366E+03 7.1409E+00 −1.9627E−05 2
Benzene 101 278.7 − 545 < 3% 7.5117E+00 2.9468E+02 −2.794E+00
n-Hexane 101 174.6 − 406.1 < 3% −6.3276E+00 6.4E+02 −6.94E−01 5.6884E+21 −10
n-Heptane 101 180.2 − 432.2 < 3% −9.4622E+00 8.7707E+02 −2.3445E−01 1.4022E+22 −10
Ammonia 101 195.4 − 393.2 < 5% −6.743E+00 5.983E+02 −7.341E−01 −3.69E−27 10
Water 101 273.2 − 646.2 < 3% −5.2843E+01 3.7036E+03 5.866E+00 −5.879E−29 10
Oxygen 101 54.4 − 150 < 25% −4.1476E+00 9.404E+01 −1.207E+00
Nitrogen 101 63.2 − 124 < 5% 1.6004E+01 −1.8161E+02 −5.1551E+00
Hydrogen 101 14 − 33 < 5% −1.1661E+01 2.47E+01 −2.61E−01 −4.1E−16 10
Hydrogen sulfide 101 187.7 − 350 < 5% −1.0905E+01 7.6211E+02 −1.1863E−01
Carbon monoxide 101 68.2 − 131.4 < 5% −4.9735E+00 9.767E+01 −1.1088E+00
Carbon dioxide 101 216.6 − 303.2 < 10% 1.8775E+01 −4.0292E+02 −4.6854E+00 −6.9171E−26 10
Helium 101 2.2 − 5.1 < 10% −9.6312E+00 −3.841E+00 −1.458E+00 −1.065E−08 10
Methanol 101 175.5 − 337.8 < 5% −2.5317E+01 1.7892E+03 2.069E+00
Ethylene glycol 101 260.2 − 576 < 5% −2.0515E+01 2.4685E+03 1.2435E+00 2.4998E+12 −5
Diethylene glycol 101 262.7 − 595.7 < 10% 1.3011E+01 2.6481E+02 −3.4184E+00 4.843E+12 −5
Triethylene Glycol 101 266 − 615.6 < 10% −2.7963E+01 3.225E+03 2.2792E+00 1.8277E+17 −7
Copyright 2005 by the Design Institute for Physical Properties (DIPPR), American Institute of Chemical Engineers (AIChE), and reproduced by permission of AIChE.
© 2006 by Taylor and Francis Group, LLC
© 2006 by Taylor and Francis Group, LLC
Appendix B 369
TABLE B.8
Vapor Viscosity at Low Pressure (Computed value has units of Pa-s)
Compound
Equation
Number
Temperature
Range
(K) Quality A B C D
Methane 102 90.7 − 1000 < 3% 5.2546E−07 5.9006E−01 1.0567E+02
Ethylene 102 169.4 − 1000 < 5% 2.0789E−06 4.163E−01 3.527E+02
Ethane 102 90.4 − 1000 < 5% 2.5906E−07 6.7988E−01 9.8902E+01
Propane 102 85.5 − 1000 < 3% 4.9054E−08 9.0125E−01
Isobutane 102 150 − 1000 < 5% 1.0871E−07 7.8135E−01 7.0639E+01
n-Butane 102 134.9 − 1000 < 3% 3.4387E−08 9.4604E−01
Isopentane 102 150 − 1000 < 5% 2.4344E−08 9.7376E−01 −9.1597E+01 1.872E+04
n-Pentane 102 143.4 − 1000 < 5% 6.3412E−08 8.4758E−01 4.1718E+01
Benzene 102 278.7 − 1000 < 3% 3.134E−08 9.676E−01 7.9E+00
n-Hexane 102 177.8 − 1000 < 5% 1.7514E−07 7.0737E−01 1.5714E+02
n-Heptane 102 182.6 − 1000 < 5% 6.672E−08 8.2837E−01 8.5752E+01
Ammonia 102 195.4 − 1000 < 10% 4.1855E−08 9.806E−01 3.08E+01
Water 102 273.2 − 1073.2 < 1% 1.7096E−08 1.1146E+00
Oxygen 102 54.4 − 1500 < 5% 1.101E−06 5.634E−01 9.63E+01
Nitrogen 102 63.2 − 1970 < 3% 6.5592E−07 6.081E−01 5.4714E+01
Hydrogen 102 14 − 3000 < 10% 1.797E−07 6.85E−01 −5.9E−01 1.4E+02
Hydrogen
sulfide
102 250 − 480 < 3% 3.9314E−08 1.0134E+00
Carbon-
monoxide
102 68.2 − 1250 < 5% 1.1127E−06 5.338E−01 9.47E+01
Carbon-
dioxide
102 194.7 − 1500 < 5% 2.148E−06 4.6E−01 2.9E+02
Helium 102 20 − 2000 < 3% 3.253E−07 7.162E−01 −9.6E+00 1.07E+02
Methanol 102 240 − 1000 < 10% 3.0663E−07 6.9655E−01 2.05E+02
Ethylene-
glycol
102 260.2 − 1000 < 10% 8.6706E−08 8.3923E−01 7.5512E+01
Diethylene-
glycol
102 262.7 − 1000 < 10% 6.7384E−08 8.489E−01 7.1139E+01
Triethylene-
Glycol
102 266 − 1000 < 10% 5.4291E−08 8.6024E−01 6.592E+01
Copyright 2005 by the Design Institute for Physical Properties (DIPPR), American Institute of Chemical Engineers
(AIChE), and reproduced by permission of AIChE.
3
7
0
F
u
n
d
a
m
e
n
t
a
l
s

o
f

N
a
t
u
r
a
l

G
a
s

P
r
o
c
e
s
s
i
n
g
TABLE B.9
Liquid Thermal Conductivity (Computed value has units of W/m-K)
Compound Equation Number Temperature Range (K) Quality A B C D E
Methane 100 90.7 − 180 < 5% 4.1768E−01 −2.4528E−03 3.5588E−06
Ethylene 100 104 − 280 < 5% 4.194E−01 −1.591E−03 1.306E−06
Ethane 100 90.4 − 300 < 3% 3.5758E−01 −1.1458E−03 6.1866E−07
Propane 100 85.5 − 350 < 3% 2.6755E−01 −6.6457E−04 2.774E−07
Isobutane 100 113.5 − 400 < 10% 2.0455E−01 −3.6589E−04
n-Butane 100 134.9 − 400 < 5% 2.7349E−01 −7.1267E−04 5.1555E−07
Isopentane 100 113.2 − 368.1 < 10% 2.1246E−01 −3.3581E−04
n-Pentane 100 143.4 − 445 < 5% 2.537E−01 −5.76E−04 3.44E−07
Benzene 100 278.7 − 413.1 < 3% 2.3444E−01 −3.0572E−04
n-Hexane 100 177.8 − 370 < 3% 2.2492E−01 −3.533E−04
n-Heptane 100 182.6 − 371.6 < 3% 2.15E−01 −3.03E−04
Ammonia 100 195.4 − 400 < 10% 1.169E+00 −2.314E−03
Water 100 273.2 − 633.2 < 1% −4.32E−01 5.7255E−03 −8.078E−06 1.861E−09
Oxygen 100 60 − 150 < 5% 2.741E−01 −1.38E−03
Nitrogen 100 63.2 − 124 < 3% 2.654E−01 −1.677E−03
Hydrogen 100 14 − 31 < 10% −9.17E−02 1.7678E−02 −3.82E−04 −3.3324E−06 1.0266E−07
Hydrogen sulfide 100 193.2 − 292.4 < 5% 4.842E−01 −1.184E−03
Carbon monoxide 100 68.2 − 125 < 3% 2.855E−01 −1.784E−03
Carbon dioxide 100 216.6 − 300 < 3% 4.406E−01 −1.2175E−03
Helium 100 2.2 − 4.8 < 3% −1.3833E−02 2.2913E−02 −5.4872E−03 4.585E−04
Methanol 100 175.5 − 337.8 < 3% 2.837E−01 −2.81E−04
Ethylene glycol 100 260.2 − 470.4 < 5% 8.8067E−02 9.4712E−04 −1.3114E−06
Diethylene glycol 100 262.7 − 518 < 5% 6.4277E−02 7.8259E−04 −1.0562E−06
Triethylene Glycol 100 266 − 561.5 < 5% 1.0753E−01 5.0392E−04 −7.2763E−07
Copyright 2005 by the Design Institute for Physical Properties (DIPPR), American Institute of Chemical Engineers (AIChE), and reproduced by permission of AIChE.
© 2006 by Taylor and Francis Group, LLC
A
p
p
e
n
d
i
x

B
3
7
1
TABLE B.10
Vapor Thermal Conductivity at Low Pressure (Computed value has units of W/m-K)
Compound Equation Number Temperature Range (K) Quality A B C D
Methane 102 111.6 − 600 < 5% 8.3983E−06 1.4268E+00 −4.9654E+01
Ethylene 102 170 − 590.9 < 5% 8.6806E−06 1.4559E+00 2.9972E+02 −2.9403E+04
Ethane 102 184.6 − 1000 < 5% 7.3869E−05 1.1689E+00 5.0073E+02
Propane 102 231.1 − 1000 < 5% −1.12E+00 1.0972E−01 −9.8346E+03 −7.5358E+06
Isobutane 102 261.4 − 1000 < 10% 8.9772E−02 1.8501E−01 6.3923E+02 1.1147E+06
n-Butane 102 272.6 − 1000 < 5% 5.1094E−02 4.5253E−01 5.4555E+03 1.9798E+06
Isopentane 102 273.2 − 1000 < 5% 8.968E−04 7.742E−01 4.56E+02 2.3064E+05
n-Pentane 102 273.2 − 1000 < 5% −6.844E+02 7.64E−01 −1.055E+09
Benzene 102 339.2 − 1000 < 5% 1.652E−05 1.3117E+00 4.91E+02
n-Hexane 102 339.1 − 1000 < 5% −6.505E+02 8.053E−01 −1.4121E+09
n-Heptane 102 339.2 − 1000 < 5% −7.0028E−02 3.8068E−01 −7.0499E+03 −2.4005E+06
Ammonia 102 200 − 900 < 5% 9.6608E−06 1.3799E+00
Water 102 273.2 − 1073.2 < 1% 6.2041E−06 1.3973E+00
Oxygen 102 80 − 2000 < 10% 4.4994E−04 7.456E−01 5.6699E+01
Nitrogen 102 63.2 − 2000 < 3% 3.3143E−04 7.722E−01 1.6323E+01 3.7372E+02
Hydrogen 102 22 − 1600 < 5% 2.653E−03 7.452E−01 1.2E+01
Hydrogen sulfide 102 212.8 − 600 < 3% 1.381E−07 1.8379E+00 −3.5209E+02 4.6041E+04
Carbon monoxide 102 70 − 1500 < 5% 5.9882E−04 6.863E−01 5.713E+01 5.0192E+02
Carbon dioxide 102 194.7 − 1500 < 10% 3.69E+00 −3.838E−01 9.64E+02 1.86E+06
Helium 102 30 − 2000 < 5% 2.26E−03 7.305E−01 −1.863E+01 4.4E+02
Methanol 102 273 − 684.4 < 5% 5.7992E−07 1.7862E+00
Ethylene glycol 102 470.4 − 1000 < 10% −8.1458E+06 −3.0502E−01 1.8325E+09 −1.1842E+13
Diethylene glycol 102 518 − 1000 < 10% 2.0395E+03 9.0063E−01 1.2238E+10
Triethylene Glycol 102 269.3 − 1000 < 25% 1.8738E−05 1.238E+00 4.7673E+02
Copyright 2005 by the Design Institute for Physical Properties (DIPPR), American Institute of Chemical Engineers (AIChE), and reproduced by permission of AIChE.
© 2006 by Taylor and Francis Group, LLC
3
7
2
F
u
n
d
a
m
e
n
t
a
l
s

o
f

N
a
t
u
r
a
l

G
a
s

P
r
o
c
e
s
s
i
n
g
TABLE B.11
Surface Tension (Computed value has units of N/m)
Compound EquationNumber Temperature Range (K) Quality A B C D
Methane 106 90.7 – 190.6 < 3% 3.6557E−02 1.1466E+00
Ethylene 106 104 – 282.3 < 3% 5.294E−02 1.278E+00
Ethane 106 90.4 – 305.3 < 3% 4.8643E−02 1.1981E+00
Propane 106 85.5 – 369.8 < 3% 5.092E−02 1.2197E+00
Isobutane 106 113.5 – 407.8 < 3% 5.1359E−02 1.2532E+00
n-Butane 106 134.9 – 425.1 < 3% 5.196E−02 1.2181E+00
Isopentane 106 113.2 – 460.4 < 3% 5.0876E−02 1.2066E+00
n-Pentane 106 143.4 – 469.7 < 3% 5.202E−02 1.2041E+00
Benzene 106 278.7 – 562 < 3% 7.1815E−02 1.2362E+00
n-Hexane 106 177.8 – 507.6 < 3% 5.5003E−02 1.2674E+00
n-Heptane 106 182.6 – 540.2 < 3% 5.4143E−02 1.2512E+00
Ammonia 106 195.4 – 405.6 < 3% 1.0162E−01 1.216E+00
Water 106 273.2 – 647.1 < 1% 1.8548E−01 2.717E+00 −3.554E+00 2.047E+00
Oxygen 106 54.4 – 154.6 < 3% 3.8066E−02 1.2136E+00
Nitrogen 106 63.2 – 126.2 < 3% 2.901E−02 1.2485E+00
Hydrogen 106 14 – 33.2 < 1% 5.345E−03 1.0646E+00
Hydrogen sulfide 106 187.7 – 373.5 < 5% 7.4256E−02 1.2997E+00
Carbon monoxide 106 68.2 – 132.9 < 3% 2.7959E−02 1.133E+00
Carbon dioxide 106 216.6 – 304.2 < 10% 8.071E−02 1.2662E+00
Helium 106 2.2 – 5.2 < 5% 5.1136E−04 1.003E+00
Methanol 100 273.1 – 503.2 < 1% 3.513E−02 −7.04E−06 −1.216E−07
Ethylene glycol 100 260.2 – 470.4 < 1% 7.4516E−02 −8.9E−05 −3.9465E−17
Diethylene glycol 100 293.2 – 518.2 < 1% 6.761E−02 −4.62E−05 −6.46E−08
Triethylene Glycol 100 293.2 – 548.2 < 1% 6.7901E−02 −6.9536E−05 −2.4419E−08
Copyright 2005 by the Design Institute for Physical Properties (DIPPR), American Institute of Chemical Engineers (AIChE), and reproduced
by permission of AIChE.
© 2006 by Taylor and Francis Group, LLC
© 2006 by Taylor and Francis Group, LLC
Appendix B 373
TABLE B.12
Equation Number Empirical Equation
100 Y A + BT + CT
2
+ DT
3
+ ET
4
101 Y Exp[A + B/T + CLn T + DT
E
]
102
105
106
, where T
r
T/T
c
107
114
, where t (1−T
r
)
116
, where t (1−T
r
)
Y
AT
C
T
D
T
B

+ + 1
2
Y
A
B
T
C
D

+ −

]
]
]
]
1 1 ( )
Y A T
r
B CT DT ET
r r r
+
+ + +
( )
[ ] 1
2 3
Y A B
C
T
C
T
D
E
T
+
|
.

`
}

]
]
]
]
]
]
]
]
+
sinh cos
2
hh
E
T
|
.

`
}

]
]
]
]
]
]
]
]
2
Y
A
t
B ACt ADt
C t CDt D t
+ − − − − −
2
2
2 3 4 2 5
2
3 2 5
Y A Bt ct dt et + + + +
0 35 2 3 4 3 . / /
Equations Noted in Tables B.2 to B.11
3
7
4
F
u
n
d
a
m
e
n
t
a
l
s

o
f

N
a
t
u
r
a
l

G
a
s

P
r
o
c
e
s
s
i
n
g
TABLE B.13
Molar Ideal Gas Heat Capacities for Various Gases (Btu/(lb-mol-؇R))

Temperature (°F)
Gas Formula Mass 0 50 60 100 150 200 250 300
Methane CH
4
16.04 8.21 8.40 8.44 8.64 8.94 9.27 9.63 10.01
Ethylene C
2
H
4
28.05 9.32 9.91 10.04 10.56 11.23 11.90 12.57 13.22
Ethane C
2
H
6
30.07 11.31 12.07 12.23 12.90 13.77 14.64 15.50 16.35
Propane C
3
H
8
44.10 15.50 16.81 17.08 18.19 19.59 20.98 22.32 23.61
Isobutane C
4
H
10
58.12 20.19 22.02 22.40 23.94 25.87 27.75 29.56 31.28
n-Butane C
4
H
10
58.12 20.89 22.55 22.89 24.31 26.11 27.88 29.61 31.26
Isopentane C
5
H
12
72.15 24.57 27.00 27.50 29.47 31.89 34.21 36.41 38.50
n-Pentane C
5
H
12
72.15 25.47 27.46 27.89 29.62 31.84 34.05 36.21 38.30
Benzene C
6
H
6
78.11 16.32 18.43 18.86 20.59 22.73 24.79 26.74 28.59
n-Hexane C
6
H
14
86.18 30.10 32.49 33.00 35.10 37.79 40.47 43.10 45.62
n-Heptane C
7
H
16
100.21 34.84 37.64 38.24 40.69 43.83 46.96 50.00 52.93
Ammonia NH
3
17.03 8.24 8.40 8.43 8.58 8.79 9.00 9.23 9.45
Air 28.96 6.94 6.95 6.95 6.96 6.97 6.99 7.01 7.03
Water H
2
O 18.02 7.99 8.01 8.01 8.03 8.07 8.12 8.17 8.23
Oxygen O
2
32.00 6.97 6.99 7.00 7.02 7.07 7.11 7.17 7.23
Nitrogen N
2
28.01 6.95 6.96 6.96 6.96 6.96 6.97 6.98 7.00
Hydrogen H
2
2.02 6.80 6.85 6.86 6.89 6.93 6.95 6.97 6.99
Hydrogen sulfide H
2
S 34.08 8.06 8.13 8.14 8.20 8.29 8.38 8.48 8.59
Carbon monoxide CO 28.01 6.95 6.96 6.96 6.96 6.97 6.98 7.00 7.03
Carbon dioxide CO
2
44.01 8.34 8.71 8.78 9.05 9.36 9.65 9.90 10.13
Methanol CH
3
OH 32.04 9.94 10.29 10.37 10.71 11.19 11.71 12.24 12.78
© 2006 by Taylor and Francis Group, LLC
Pure fluid values computed by application of coefficients given in Table B.6. Values for air taken from Engineering Data Book (2004b).
A
p
p
e
n
d
i
x

B
3
7
5
TABLE B.14
Molar Ideal Gas Heat Capacities for Various Gases [kJ/(kmol-K)]

Temperature (؇C)

Gas Formula Mass -25 0 10 25 50 75 100 125 150
Methane CH
4
16.043 34.20 34.84 35.15 35.68 36.70 37.87 39.15 40.53 41.96
Ethylene C
2
H
4
28.054 38.41 40.57 41.50 42.95 45.44 47.97 50.50 53.00 55.43
Ethane C
2
H
6
30.07 46.58 49.34 50.53 52.39 55.60 58.88 62.16 65.40 68.58
Propane C
3
H
8
44.097 63.57 68.34 70.36 73.47 78.75 84.02 89.20 94.21 99.04
Isobutane C
4
H
10
58.124 82.63 89.36 92.18 96.51 103.79 111.00 118.02 124.77 131.25
n-Butane C
4
H
10
58.124 85.80 91.83 94.39 98.35 105.08 111.84 118.49 124.94 131.17
Isopentane C
5
H
12
72.151 100.31 109.35 113.06 118.65 127.89 136.86 145.47 153.68 161.51
n-Pentane C
5
H
12
72.151 104.66 111.88 114.98 119.80 128.07 136.45 144.74 152.84 160.69
Benzene C
6
H
6
78.114 66.13 73.92 77.14 82.03 90.17 98.12 105.78 113.07 119.99
n-Hexane C
6
H
14
86.178 123.66 132.30 136.04 141.84 151.87 162.04 172.12 181.94 191.41
n-Heptane C
7
H
16
100.205 143.07 153.23 157.61 164.41 176.12 187.97 199.69 211.09 222.08
Ammonia NH
3
17.031 34.36 34.90 35.15 35.55 36.29 37.09 37.92 38.76 39.61
Air 28.964 29.11 29.12 29.12 29.13 29.14 29.17 29.20 29.25 29.31
Water H
2
O 18.015 33.43 33.49 33.52 33.58 33.70 33.85 34.03 34.23 34.45
Oxygen O
2
31.999 29.18 29.25 29.28 29.35 29.48 29.65 29.84 30.06 30.29
Nitrogen N
2
28.013 29.11 29.12 29.12 29.13 29.14 29.17 29.20 29.25 29.31
Hydrogen H
2
2.016 28.41 28.60 28.68 28.78 28.92 29.04 29.13 29.20 29.25
Hydrogen sulfide H
2
S 34.076 33.69 33.92 34.02 34.19 34.50 34.84 35.20 35.58 35.97
Carbon monoxide CO 28.01 29.11 29.12 29.13 29.14 29.16 29.20 29.26 29.33 29.42
Carbon dioxide CO
2
44.01 34.51 35.91 36.45 37.24 38.48 39.62 40.65 41.58 42.44
Methanol CH
3
OH 32.04 41.29 42.51 43.08 44.01 45.72 47.58 49.55 51.57 53.61
© 2006 by Taylor and Francis Group, LLC
Pure fluid values computed using coefficients given in Table B.6. Values for air taken from Engineering Data Book (2004b).
© 2006 by Taylor and Francis Group, LLC
376 Fundamentals of Natural Gas Processing
B.3.3 PHYSICAL PROPERTIES OF ALKANOLAMINES
AND ALKANOLAMINE-WATER SYSTEMS
TABLE B.15
Selected Physical Properties of Alkanolamines
Compound MEA
a
DEA
a
TEA
a
MDEA
b
Molecular weight 61.08 15.14 149.19 119.16
Specific gravity at 20/20°C
∆Sp. Gr./∆t (°C
−1
)
1.017
0.00080
1.092
0.00065
1.126
0.00059
1.041
0.00076
Boiling point, °C at 760 mm Hg
At 50 mm Hg, °C
At 10 mm Hg, °C
170.4
101
71
268
182
150
335
245
205
247.3
163.5
128.6
Vapor pressure at 20°C, mm Hg <1 <0.01 <0.001 <0.01
Freezing point, °C (°F), (pour point) 10.5 (50.9) 28.0 (82.4) 21.6 (70.9) −21 (−6)
Viscosity, cP
At 20°C
At 40°C

24.1

921

101
33.8
Heat of combustion, Btu/lb (cal/g)
at 25°C
−12,200
(−6780)
Flash point, Tag closed cup
(ASTM D56), °C (°F)
96 (205) 191 (375) 208 (407) 138 (280)
c
a
Dow Chemical (2003b). Ethanolamine brochure
b
Dow Chemical (2003a).
c
Pensky-Martens Closed Cup (ASTM D93).
FIGURE B.1 Densities of monoethanolamine (MEA) – water mixtures as a function of composition
and temperature. (Adapted from Dow Chemical, 2003b.)
100 80 60 40 20 0
7.8
8.0
8.2
8.4
D
e
n
s
i
t
y
,

l
b
/
g
a
l
8.6
8.8
9.0
MEA concentration, Wt%
210°F
180°F
150°F
120°F
90°F
60°F
30°F
0°F
Freezing point curve
© 2006 by Taylor and Francis Group, LLC
Appendix B 377
FIGURE B.2 Densities of diethanolamine (DEA) – water mixtures as a function of com-
position and temperature. (Adapted from Dow Chemical, 2003b.)
FIGURE B.3 Densities of triethanolamine (TEA) – water mixtures as a function of com-
position and temperature. (Adapted from Dow Chemical, 2003b.)
100 80 60 40 20 0
8.0
8.2
8.4
8.6
D
e
n
s
i
t
y
,

l
b
/
g
a
l 8.8
9.0
9.2
9.4
DEA concentration, Wt%
Freezing point curve
0°F
−20°F
Boiling point
210°F
180°F
150°F
120°F
90°F
60°F
30°F
240°F
100 80 60 40 20 0
7.8
8.0
8.2
D
e
n
s
i
t
y
,

l
b
/
g
a
l
8.4
8.6
8.8
9.0
9.2
TEA concentration, Wt%
9.4
9.6
240°F
Boiling point
120°F
210°F
180°F
150°F
90°F
60°F
30°F
Freezing point curve
0°F
© 2006 by Taylor and Francis Group, LLC
378 Fundamentals of Natural Gas Processing
FIGURE B.4 Densities of methyl diethanolamine (MDEA) – water mixtures as a function
of composition and temperature. (Adapted through the courtesy of INEOS Oxide).
FIGURE B.5 Heat capacities of monoethanolamine (MEA) – water mixtures as a function
of composition and temperature. (Adapted from Dow Chemical, 2003b.)
140 100
100 Wt%
MDEA
80 60
Temperature, °C
Temperature, °F
40 20 0
8.0
8.1
8.2
8.3
8.4
8.5
8.6
8.7
8.8
248 212 176 140 104 32 68
120
50 Wt%
40 Wt%
284
0.959
0.971
0.983
0.995
1.007
1.019
1.031
1.042
1.054
D
e
n
s
i
t
y
,

l
b
/
g
a
l
D
e
n
s
i
t
y
,

g
m
/
c
m
3
120 100 60
Temperature, °C
Temperature, °F
40 20 0
0.6
0.7
0.8
H
e
a
t

c
a
p
a
c
i
t
y
,

B
t
u
/
l
b
-
°
F
0.9
1.0
1.1
80
40 80 120 160 180 240
0
Wt% MEA
25
50
75
100
© 2006 by Taylor and Francis Group, LLC
Appendix B 379
FIGURE B.6 Heat capacities of diethanolamine (DEA) – water mixtures as a function of
composition and temperature. (Adapted from Dow Chemical, 2003b.)
FIGURE B.7 Heat capacities of triethanolamine (TEA) – water mixtures as a function of
composition and temperature. (Adapted from Dow Chemical, 2003b.)
120 100 60 40 20 0
0.6
0.7
0.8
0.9
1.0
1.1
80
40 80 120 160 180 240
0
25
50
75
100
Temperature, °C
Temperature, °F
H
e
a
t

c
a
p
a
c
i
t
y
,

B
t
u
/
l
b
-
°
F
Wt% DEA
120 100 60 40 20 0
0.6
0.7
0.8
0.9
1.0
1.1
80
40 80 120 160 180 240
0
25
50
75
100
0.5
Temperature, °C
Temperature, °F
H
e
a
t

c
a
p
a
c
i
t
y
,

B
t
u
/
l
b
-
°
F
Wt% TEA
© 2006 by Taylor and Francis Group, LLC
380 Fundamentals of Natural Gas Processing
FIGURE B.8 Heat capacities of methyldiethanolamine (MDEA) – water mixtures as a
function of composition and temperature. (Adapted through the courtesy of INEOS Oxide).
FIGURE B.9 Viscosities of monoethanolamine (MEA) – water mixtures as a function of
composition and temperature. (Adapted from Dow Chemical, 2003b.)
140 100 80 60 40 20 0
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
248 212 176 140 104 32 68
120
284
100
50
40
Temperature, °C
Temperature, °F
S
p
e
c
i

c

h
e
a
t
,

B
t
u
/
l
b
-
°
F
Wt% MDEA
100 80 60 40 20 0
MEA concentration, Wt%
0.1
0.2
0.4
0.6
1
V
i
s
c
o
s
i
t
y
,

c
P
2
4
6
10
20
40
60
100
30°C (86°F)
75°C (167°F)
100°C (212°F)
50°C (122°F)
© 2006 by Taylor and Francis Group, LLC
Appendix B 381
FIGURE B.10 Viscosities of diethanolamine (DEA) – water mixtures as a function of
composition and temperature. (Adapted from Dow Chemical, 2003b.) Reference implies
the viscosity of TEA−water mixtures.
FIGURE B.11 Viscosities of methyldiethanolamine (MDEA) – water mixtures as a func-
tion of composition and temperature. (Adapted through the courtesy of INEOS Oxide).
100 80 60 40 20
DEA concentration, Wt%
0
0.1
0.2
0.4
0.6
1
2
4
6
10
20
40
60
100
V
i
s
c
o
s
i
t
y
,

c
P
200
400
1000
600
30°C (86°F)
50°C (122°F)
75°C (167°F)
100°C (212°F)
100
40
50
120
100 80 60 40 20 0
0
10
20
30
40
50
60
70
80
Temperature, °C
Temperature, °F
248 212 176 140 104 32 68
V
i
s
c
o
s
i
t
y
,

c
P
Wt% MDEA
that DEA and TEA viscosities are sufficiently close that this figure can be used to estimate
© 2006 by Taylor and Francis Group, LLC
382 Fundamentals of Natural Gas Processing
B.3.4 PHYSICAL PROPERTIES OF GLYCOLS AND GLYCOL-WATER
SYSTEMS
TABLE B.16
Selected Properties of Glycols Used for Dehydration
Ethylene
Glycol
a
Diethylene
Glycol
b
Triethylene
Glycol
c
Tetraethylene
Glycol
d
Molar mass, lb/lb-mol, kg/kg-mol 62.07 106.12 150.17 194.23
Normal boiling point, °F (°C) 386.8
(197.1)
473.5 (245.3) 550.4 (288.0) 625.5 (329.7)
(decomposes)
Vapor pressure at 20°C, psia (kPa) 0.00012 (0.0075) 4×10−5
(0.0003)
< 1×10−5
(<0.001)
< 1×10−5
(<0.001)
Specific gravity (20/20°C) 1.1153 1.1182 1.1255 1.1247
∆ Specific gravity/∆T (10 to 40°C),
°F
−1
(°C
−1
)
0.00039 (0.00070) 0.00040
(0.00073)
0.00043
(0.00078)
0.00044
(0.00080)
Viscosity at 20°C, cP (mPs) ~25
(~25)
35.7 (35.7) 49.0 (49.0) 58.3 (58.3)
Critical pressure,psia (bar) 1190 (82) 668 (46.05) 480 (33.1) 375 (25.9)
Critical specific volume, ft
3
/lb-mol
(m
3
/kmol)
3.06 (0.191) 5.00 (0.312) 7.10 (0.443) 9.03 (0.564)
Critical temperature, °F (°C) 836.33 (446.85) 764.33
(406.85)
824(440) 971.6 (522)
Onset of initial decomposition,
°F (°C)
464 (240) 464 (240) 464 (240) 464 (240)
Autoignition temperature, °F (°C) 801 (427) 687(364) 660 (349) 676 (358)
Flammable limits in air, lower,
vol%
2.0 0.9
Flammable limits in air, upper,
vol%
12.3 9.2
a
MEGlobal (2005b).
b
MEGlobal (2005a).
c
Dow Chemical (2003d).
d
Dow Chemical (2003c).
© 2006 by Taylor and Francis Group, LLC
Appendix B 383
The following solutes are reported to be completely miscible with all of the above
glycols:
Water, Methanol, Monoethanolamine, Diethanolamine
TABLE B.17
Selected Solubility for Compounds in Glycols at 25؇C (g solute/100 ml
of glycol)
Solute
Ethylene
Glycol
a
Diethylene
Glycol
b
Triethylene
Glycol
c
Tetraethylene
Glycol
d
Heptane Slightly Soluble 0.03 Slightly Soluble Slightly Soluble
Benzene 6.0 45.5 Completely miscible Completely
miscible
Toluene 3.1 20.7 33 89
a
MEGlobal (2005b).
b
MEGlobal (2005a).
c
Dow Chemical (2003d).
d
Dow Chemical (2003c).
TABLE B.18
Vapor Pressure Equations for Glycols
Ethylene
Glycol
a
Diethylene
Glycol
b
Triethylene
Glycol
c
Tetraethylene
Glycol
d
A 8.21211 7.7007954 7.6302007 7.3101
B 2161.91 2019.2548 2156.4581 2076.0
C 208.43 173.66153 165.92442 139.0
Range of
applicability, °C
20 to 198 10 to 250 150 to 300 190 to 330
The coefficients are for the equation log
10
(P [mm Hg]) A – B/(C + t [°C]).
a
MEGlobal (2005b).
b
MEGlobal (2005a).
c
Dow Chemical (2003d).
d
Dow Chemical (2003c).
© 2006 by Taylor and Francis Group, LLC
384 Fundamentals of Natural Gas Processing
FIGURE B.12 Specific gravities of ethylene glycol−water mixtures as a function of con-
centration and temperature. (Courtesy of MEGlobal.)
FIGURE B.13 Specific gravities of diethylene glycol−water mixtures as a function of
concentration and temperature. (Courtesy of MEGlobal.)
Freezing point curves
−50°F
0°F
50°F
100°F
1.18
1.13
1.08
1.03
A
p
p
a
r
e
n
t

s
p
e
c
i

c

g
r
a
v
i
t
y
,

t
/
6
0
°
F
0.98
0.93
0.88
0 10 20 30 40 50
Ethylene glycol, % by weight in water
60 70 80 90 100
150°F
200°F
250°F
300°F
350°F
1.20
1.15
1.10
1.05
A
p
p
a
r
e
n
t

s
p
e
c
i

c

g
r
a
v
i
t
y
,

t
/
6
0
°
F
1.00
0.95
0.90
0 40 20
Diethylene glycol, percent by weight in water
60 80 100
−50°F
0°F
50°F
100°F
150°F
200°F
250°F
300°F
Freezing point curve
© 2006 by Taylor and Francis Group, LLC
Appendix B 385
FIGURE B.14 Specific gravities of triethylene glycol−water mixtures as a function of
concentration and temperature. (Courtesy of Dow Chemical.)
FIGURE B.15 Specific gravities of tetraethylene glycol−water mixtures as a function of
concentration and temperature. (Courtesy of Dow Chemical.)
1.18
1.16
1.14
1.12
1.10
1.08
1.06
1.04
1.02
1.00
0.98
0.96
A
p
p
a
r
e
n
t

s
p
e
c
i

c

g
r
a
v
i
t
y
,

t
/
6
0
°
F
0.94
0.92
0.90
0 40 50 20 30 10
Triethylene glycol, percent by weight in water
60 70 80 90 100
−50°F
0°F
50°F
100°F
150°F
200°F
250°F
300°F
Freezing point
curves
1.20
1.18
1.16
1.14
1.12
1.10
1.08
1.06
1.04
1.02
1.00
0.98
0.96
A
p
p
a
r
e
n
t

s
p
e
c
i

c

g
r
a
v
i
t
y
,

t
/
6
0
°
F
0.94
0.92
0.90
0 40 50 20 30 10
Tetraethylene glycol, percent by weight in water
60 70 80 90 100
50°F
Freezing point curves
0°F
5
0
°F
1
0
0
°F
1
5
0
°F
2
0
0
°F
2
5
0
°F
3
0
0
°F
© 2006 by Taylor and Francis Group, LLC
386 Fundamentals of Natural Gas Processing
FIGURE B.16 Specific heats of ethylene glycol−water mixtures as a function of concen-
tration and temperature. (Courtesy of MEGlobal.)
FIGURE B.17 Specific heats of diethylene glycol−water mixtures as a function of con-
centration and temperature. (Courtesy of MEGlobal.)
Freezing point
curves
0°F
10°F
20°F
−50
1.1
1.0
0.9
0.8
0.7
0.6
0.5
0.4
−30 −10 10 30 50
Temperature, °C
S
p
e
c
i

c

h
e
a
t
,

B
t
u
/
l
b
-
°
F
70 90 110 130 150
−58 −18 22 62 142 102 182
Temperature, °F
222 262 302
30°F
40°F
50°F
60°F
70°F
80°F
90°F
100°F
E
t
h
y
l
e
n
e

g
l
y
c
o
l
,

%

b
y

w
e
i
g
h
t

i
n

w
a
t
e
r
−50
1.1
1.0
0.9
0.8
0.7
0.6
0.5
0.4
50 0
Temperature, °C
S
p
e
c
i

c

h
e
a
t
,

B
t
u
/
l
b
-
°
F
100 150
−40 0 40 80 160 120 200
Temperature, °F
240 280
0
10
30
50
70
90
100
D
i
e
t
h
y
l
e
n
e

g
l
y
c
o
l
,

p
e
r
c
e
n
t

b
y

w
e
i
g
h
t

i
n

w
a
t
e
r
Freezing point curve
© 2006 by Taylor and Francis Group, LLC
Appendix B 387
FIGURE B.18 Specific heats of triethylene glycol−water mixtures as a function of con-
centration and temperature. (Courtesy of Dow Chemical.)
FIGURE B.19 Specific heats of tetraethylene glycol−water mixtures as a function of
concentration and temperature. (Courtesy of Dow Chemical.)
© 2006 by Taylor and Francis Group, LLC
388 Fundamentals of Natural Gas Processing
FIGURE B.20 Viscosities of ethylene glycol−water mixtures as a function of concentra-
tion and temperature. (Courtesy of MEGlobal.)
FIGURE B.21 Viscosities of diethylene glycol−water mixtures as a function of concen-
tration and temperature. (Courtesy of MEGlobal.)
1000
100
10
V
i
s
c
o
s
i
t
y
,

c
e
n
t
i
p
o
i
s
e
s

(
c
P
)
1
0.1
0 10 20 30 40
Ethylene glycol, % by weight in water
50 60 70 80 90 100
10°F
30°F
50°F
100°F
150°F
200°F
250°F
300°F
350°F
−50°F
−40°F
−30°F
−20°F
−10°F
0°F
Freezing
point
curve
Freezing point
curve